[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$ffFUljuElgdP-i1mZ4wzexYr4TTvUKDDrmCNWTaL4Fwo":3,"$f-gwIXJgDGgGv9bmRu29vUWmMAqDVQVCWspFuwnTv5fs":28,"$fs8386zdAUI-Rj5TzVbzTerjIHduHTXuIcIJmyYrxqs0":127,"$fBuPzgYSZ2qvCIvMieT5y4oi-25q1ef7qJGzWFioAT14":300,"$ffiTUArCIusuXy_72zAFgwM8K8JtnSKnJ0pN_ASE6qSc":797,"notifications-unread":1707,"$fcnp-1s7Afu5nMbzP0-SpadvNYh92tK-LNEathDIll5Q":2255},[4,9,14,19,24],{"id":5,"slug":6,"name":7,"color_token":8},"5878a593-ac48-4ff3-a25b-35a7de448daa","narzedzia-ai","Narzędzia AI","accent",{"id":10,"slug":11,"name":12,"color_token":13},"ec013b2c-e6c4-4c31-b1ec-9587c629a169","praktyczne-zastosowania","Praktyczne zastosowania","gold",{"id":15,"slug":16,"name":17,"color_token":18},"165808b7-bea9-4eef-8280-014b3d85e6bf","news-analizy","News & analizy","azure",{"id":20,"slug":21,"name":22,"color_token":23},"d6c684e6-7b4c-40d9-907b-469f30dc2126","tutoriale-how-to","Tutoriale how-to","olive",{"id":25,"slug":26,"name":27,"color_token":8},"73506ae5-4a59-4ce3-9712-58fbb8565fc4","nauka-biznes-marketing","Nauka biznes & marketing",[29,34,39,44,49,53,58,63,68,72,77,82,87,92,97,102,107,112,117,122],{"slug":30,"draft_id":31,"pillar_id":33},"google-vs-anthropic-kto-naprawde-kontroluje-open-source-ai",{"title":32},"Google vs Anthropic: kto naprawdę kontroluje open source AI?",{"name":17},{"slug":35,"draft_id":36,"pillar_id":38},"78-deklaruje-5-mierzy-gdzie-reszta-gubi-pieniadze",{"title":37},"78% deklaruje, 5% mierzy. Gdzie reszta gubi pieniądze?",{"name":27},{"slug":40,"draft_id":41,"pillar_id":43},"27-razy-zbudowalem-tego-samego-bota-oto-co-naprawde-psuje-sie-po-wyjsciu-z-demo",{"title":42},"Zapier, Make czy n8n: agenci AI i automatyzacja bez kodu w 2026",{"name":7},{"slug":45,"draft_id":46,"pillar_id":48},"news-analizy-nawigacja-w-dynamicznym-swiecie-ai",{"title":47},"Newsy i analizy AI w 2026: jak odróżnić zmianę realną od szumu",null,{"slug":50,"draft_id":51,"pillar_id":48},"praktyczne-zastosowania-ai-od-pilotazu-do-produkcji",{"title":52},"Praktyczne zastosowania AI: jak przejść od pilotażu do produkcji",{"slug":54,"draft_id":55,"pillar_id":57},"90-inwestuje-12-mierzy-to-jest-roznica-miedzy-zyskiem-a-wydatkiem",{"title":56},"90% inwestuje, 12% mierzy. To jest różnica między zyskiem a wydatkiem.",{"name":27},{"slug":59,"draft_id":60,"pillar_id":62},"web-search-rag-pipeline-zanim-napiszesz-kolejny-scraper-przeczytaj-o-czterech-pulapkach",{"title":61},"Twój scraper spala tokeny na menu nawigacyjnym. Oto co zrobić zamiast tego",{"name":12},{"slug":64,"draft_id":65,"pillar_id":67},"hermes-agent-vs-gpt-4-kto-naprawde-wygrywa-w-open-source",{"title":66},"Hermes Agent vs. GPT-4: Kto naprawdę wygrywa w open-source?",{"name":17},{"slug":69,"draft_id":70,"pillar_id":48},"tutoriale-how-to-praktyczne-integracje-ai-krok-po-kroku",{"title":71},"Tutoriale how-to: Praktyczne integracje AI krok po kroku",{"slug":73,"draft_id":74,"pillar_id":76},"twoj-pilot-ai-zachwycil-zarzad-dlaczego-nigdy-nie-trafi-na-hale-produkcyjna",{"title":75},"95% pilotaży AI nie trafia do produkcji — i nie chodzi o kod",{"name":12},{"slug":78,"draft_id":79,"pillar_id":81},"twoj-agent-ma-72-skutecznosci-zmiana-modelu-tego-nie-naprawi",{"title":80},"LLM jako selektor akcji: dlaczego 72% to pułap, którego nie przeskoczysz bez zmiany architektury",{"name":12},{"slug":83,"draft_id":84,"pillar_id":86},"openai-i-microsoft-kto-zyskuje-kto-traci-na-zerwaniu-wylacznosci",{"title":85},"OpenAI i Microsoft: kto zyskuje, kto traci na zerwaniu wyłączności",{"name":17},{"slug":88,"draft_id":89,"pillar_id":91},"5-dni-zero-kodu-dzialajacy-agent-ai-czy-google-naprawde-da-founderom-to-czego-potrzebuja",{"title":90},"5 dni, zero kodu, działający agent AI – czy Google naprawdę da founderom to, czego potrzebują?",{"name":22},{"slug":93,"draft_id":94,"pillar_id":96},"karpathy-w-anthropic-kto-teraz-naprawde-wyprzedzi-openai",{"title":95},"Karpathy w Anthropic: kto teraz naprawdę wyprzedzi OpenAI?",{"name":17},{"slug":98,"draft_id":99,"pillar_id":101},"chatgpt-w-rozmowach-o-depresji-czy-zwolnieniach-kiedy-ai-naprawde-rozumie-o-co-chodzi",{"title":100},"ChatGPT w rozmowach o depresji czy zwolnieniach — kiedy AI naprawdę rozumie, o co chodzi?",{"name":17},{"slug":103,"draft_id":104,"pillar_id":106},"twoj-agent-ai-wlasnie-zjadl-trucizne-i-nawet-o-tym-nie-wie",{"title":105},"Twój agent AI właśnie przeczytał stronę, która go okłamała — i nie wie o tym",{"name":12},{"slug":108,"draft_id":109,"pillar_id":111},"kiedy-brak-llm-gateway-zaczyna-kosztowac-mapa-decyzji-dla-operatora",{"title":110},"Kiedy jeden provider LLM zaczyna kosztować więcej niż cały zespół",{"name":12},{"slug":113,"draft_id":114,"pillar_id":116},"egzaminy-w-erze-ai-czy-chatgpt-zdalby-mature-lepiej-niz-twoje-dziecko",{"title":115},"Egzaminy w erze AI: Czy ChatGPT zdałby maturę lepiej niż twoje dziecko?",{"name":17},{"slug":118,"draft_id":119,"pillar_id":121},"nemotron-3-nano-vs-reszta-stawki-4b-kto-wygrywa-lokalny-benchmark-finansowy-w-2026",{"title":120},"Nemotron 3 Nano 4B: co potwierdzają źródła o modelu NVIDIA",{"name":7},{"slug":123,"draft_id":124,"pillar_id":126},"15-minut-vs-30-sekund-jak-przyspieszyc-pandasa-bez-kupowania-nowego-serwera",{"title":125},"15 minut vs 30 sekund: jak przyspieszyć Pandasa bez kupowania nowego serwera",{"name":22},[128,160,186,212,237,258,278],{"id":129,"label":130,"label_en":131,"icon":132,"subcategories":133,"count":159},"generowanie-tresci","Generowanie treści","Content Generation","pen-tool",[134,139,144,149,154],{"id":135,"label":136,"description":137,"count":138},"tekst-llm","Tekst i LLM","Chatboty, asystenci pisania, generatory copy, parafrazy",28,{"id":140,"label":141,"description":142,"count":143},"obrazy-design","Obrazy i design","Generowanie obrazów, edycja zdjęć, design graficzny",115,{"id":145,"label":146,"description":147,"count":148},"wideo-animacja","Wideo i animacja","Generowanie wideo, edycja, animacja AI",58,{"id":150,"label":151,"description":152,"count":153},"audio-mowa","Audio i mowa","TTS, voice cloning, generowanie muzyki, podcasty AI",43,{"id":155,"label":156,"description":157,"count":158},"kodowanie","Kodowanie","Asystenci kodowania, generacja kodu, code review",18,262,{"id":161,"label":162,"label_en":163,"icon":164,"subcategories":165,"count":185},"automatyzacja","Automatyzacja","Automation","zap",[166,170,175,180],{"id":167,"label":168,"description":169,"count":153},"workflow","Workflow i integracje","Automatyzacja procesów, integracje API, no-code",{"id":171,"label":172,"description":173,"count":174},"agenci-ai","Agenci AI","Autonomiczni agenci, frameworki agentowe, MCP",51,{"id":176,"label":177,"description":178,"count":179},"web-scraping","Web scraping i harvesting","Ekstrakcja danych, scraping, monitoring stron",6,{"id":181,"label":182,"description":183,"count":184},"rpa","RPA i automatyzacja biura","Robotic Process Automation, automatyzacja zadań biurowych",0,100,{"id":187,"label":188,"label_en":189,"icon":190,"subcategories":191,"count":211},"biznes-marketing","Biznes i marketing","Business & Marketing","briefcase",[192,197,202,207],{"id":193,"label":194,"description":195,"count":196},"marketing-seo","Marketing i SEO","SEO, content marketing, reklamy, social media",76,{"id":198,"label":199,"description":200,"count":201},"sprzedaz-crm","Sprzedaż i CRM","CRM AI, lead generation, sales automation",39,{"id":203,"label":204,"description":205,"count":206},"analityka","Analityka i BI","Business Intelligence, analiza danych, raportowanie",7,{"id":208,"label":209,"description":210,"count":179},"finanse-ksiegowosc","Finanse i księgowość","Automatyzacja finansów, fakturowanie, księgowość AI",128,{"id":213,"label":214,"label_en":215,"icon":216,"subcategories":217,"count":201},"produktywnosc","Produktywność","Productivity","check-square",[218,222,227,232],{"id":219,"label":220,"description":221,"count":206},"asystenci","Asystenci AI","Asystenci osobisti, AI chat, asystenci głosowi",{"id":223,"label":224,"description":225,"count":226},"notatniki-pm","Notatniki i zarządzanie projektami","AI note-taking, project management, organizacja pracy",23,{"id":228,"label":229,"description":230,"count":231},"komunikacja","Komunikacja i email","Email AI, komunikacja zespołowa, kalendarze",5,{"id":233,"label":234,"description":235,"count":236},"research-wiedza","Research i zarządzanie wiedzą","Research AI, bazy wiedzy, zarządzanie dokumentami",4,{"id":238,"label":239,"label_en":240,"icon":241,"subcategories":242,"count":257},"edukacja-nauka","Edukacja i nauka","Education & Science","book-open",[243,247,252],{"id":244,"label":245,"description":246,"count":226},"e-learning","E-learning i kursy","Platformy edukacyjne, generacja kursów, tutoring AI",{"id":248,"label":249,"description":250,"count":251},"jezyki","Nauka języków","Aplikacje do nauki języków, translatory, konwersacje AI",1,{"id":253,"label":254,"description":255,"count":256},"akademia-research","Badania naukowe","Research akademicki, analiza publikacji, cytowania",2,26,{"id":259,"label":260,"label_en":261,"icon":262,"subcategories":263,"count":277},"infrastruktura","Infrastruktura i API","Infrastructure & API","server",[264,269,273],{"id":265,"label":266,"description":267,"count":268},"api-modele","API i modele","API LLM, hosting modeli, platformy AI",19,{"id":270,"label":271,"description":272,"count":236},"mcp-serwery","MCP i narzędzia deweloperskie","Model Context Protocol, SDK, frameworki AI",{"id":274,"label":275,"description":276,"count":251},"bezpieczenstwo","Bezpieczeństwo i compliance","AI security, monitoring, compliance, governance",24,{"id":279,"label":280,"label_en":281,"icon":282,"subcategories":283,"count":299},"pozostale","Pozostałe","Other","more-horizontal",[284,289,294],{"id":285,"label":286,"description":287,"count":288},"wykrywanie-ai","Wykrywanie AI","AI content detection, deepfake detection, watermarking",13,{"id":290,"label":291,"description":292,"count":293},"rozrywka","Rozrywka i lifestyle","AI w rozrywce, randki, gry, kreatywność",10,{"id":295,"label":296,"description":297,"count":298},"inne","Inne","Nieskategoryzowane narzędzia AI",12,35,{"data":301,"_meta":794,"alternatives":795,"deals":796},{"source":302,"source_key":303,"source_page_url":304,"name":305,"slug":306,"description":307,"title":308,"category":309,"access_model":310,"pricing_evidence":311,"capability_evidence":315,"headings":326,"breadcrumbs":331,"pros":332,"cons":344,"external_links":349,"affiliate_links":411,"official_link_candidates":412,"text_excerpt":423,"fetched_at":424,"final_url":304,"http_status":425,"content_type":426,"content_hash":427,"cache_file":428,"published":429,"verification_status":430,"official_url":431,"official_evidence":432,"api_available":675,"blog_titles":676,"data_region":680,"data_training":681,"entity_type":682,"external_rating":680,"features":683,"free_summary":702,"gdpr":680,"integrations":703,"languages":719,"license":722,"limitations":723,"model_names":732,"polish_support":734,"pricing_plans":735,"pricing_summary":743,"privacy_summary":744,"release_date":745,"research_links":746,"research_summary":747,"security_summary":748,"self_hosted":675,"short_description":749,"starting_price":739,"supported_platforms":750,"target_users":757,"trial_summary":680,"use_cases":765,"vendor":781,"status":782,"last_verified_at":783,"alternatives":784,"quality":789,"category_v3":265,"category_v3_reason":791,"deep_card":792,"deep_card_generated":429,"category_v3_main":259,"category_v3_label":266,"category_v3_main_label":260,"_category_fixed":429,"evidence":793},"AIxploria","aixploria","https:\u002F\u002Fwww.aixploria.com\u002Finkling-thinking-machines\u002F","Inkling","inkling-thinking-machines","Inkling (par Thinking Machines) est un modèle IA open-weights de 975B paramètres, multimodal, avec raisonnement contrôlable et fine-tuning disponible sur Tinker.","Inkling : Avis, Prix, Info & 41 IA Alternatives | 2026 | AIxploria","audio","open_source",[312,313,314],"97 Visiter Freemium Kimi K3","248 Visiter Freemium GPT‑5.5 Instant","89 Visiter Freemium Claude Sonnet 5",[316,317,318,319,320,321,322,323,324,325],"Inkling est le premier modèle d'IA en poids ouverts de Thinking Machines Lab, le laboratoire fondé par Mira Murati, ancienne directrice technique d'OpenAI. Ses poids se téléchargent gratuitement sous licence Apache 2.0; seule l'inférence, sur vos machines ou via une API partenaire, se paie. Sous le capot, un mélange d'experts de 975 milliards de paramètres , dont 41 milliards actifs par requête, et une fenêtre de contexte d'un million de tokens. Il lit texte, images et audio, puis répond en texte, code compris.","Avantages Poids téléchargeables et modifiables sous licence Apache 2.0 Comprend nativement texte, images et audio Fenêtre de contexte d'un million de tokens Effort de réflexion réglable pour maîtriser les coûts Déclinaison Inkling-Small pour les configurations modestes Inconvénients Sorties limitées au texte, sans génération d'images Version complète réservée aux gros clusters GPU Performance brute en retrait des tout meilleurs modèles Trois sens en entrée, un curseur d'effort en sortie","Voie d'accès Ce que vous obtenez Pour qui Inkling Playground essai conversationnel dans le navigateur les curieux, sans installation Hugging Face poids complets BF16, NVFP4 et variantes GGUF équipes disposant d'un cluster GPU Tinker fine-tuning sur vos données, contexte 64K ou 256K personnalisation métier API partenaires (Together AI, Fireworks, Baseten, Modal, Databricks) inférence hébergée facturée à l'usage intégration dans vos applications Essayer Inkling gratuitement Questions fréquentes","Inkling est-il gratuit ? Oui, les poids d'Inkling se téléchargent gratuitement sur Hugging Face sous licence Apache 2.0. Le coût se déplace vers l'exécution: soit votre propre matériel, soit une API facturée à l'usage chez Together AI, Fireworks, Baseten, Modal ou Databricks. Le fine-tuning via Tinker se paie également.","Inkling est-il vraiment open source ? Techniquement, Inkling est un modèle à poids ouverts plutôt qu'open source complet: la licence Apache 2.0 couvre usage commercial, modification et redistribution sans redevance, mais les données et le code d'entraînement restent privés. C'est le même standard que les publications de Llama ou DeepSeek.","Verdict : Champion des classements, non; matière première, oui: si votre équipe cherche un modèle multimodal à affiner sur ses propres données sans dépendre d'une API fermée, Inkling est l'une des fondations ouvertes les plus ambitieuses à l'ouest du Pacifique.","« Créez rapidement des miniatures accrocheuses pour vos vidéos sur YouTube, Twitch et Facebook. Attirez plus de vues et améliorez facilement le taux de clics sur vos clip vidéos »","« Le modèle LLM de chez Moonshot qui rivalise avec les meilleurs modèles propriétaires sur de nombreux benchmarks. Conçu pour le codage longue durée, le raisonnement avancé et les tâches de knowledge work, c'est aussi le premier modèle open source à franchir la barre des 2,8 trillions de paramètres»","« Le dernier modèle d’OpenAI pour ceux qui ont besoin d’un IA plus forte en code, en science et en cybersécurité. Disponible directement dans ChatGPT ou via l'API en trois versions : Sol, Terra et Luna»","« Exploitez le modèle IA d'Anthropic, plus fiable, plus honnête et 4 fois moins enclin à laisser passer ses propres erreurs. Son mode rapide délivre jusqu'à 2,5x plus de tokens par seconde (au même tarif API que la version précédente)",[327,328,329,330],"Trois sens en entrée, un curseur d'effort en sortie","Inkling, une base à façonner plutôt qu'un trophée de podium","Où essayer et télécharger Inkling","Questions fréquentes",[],[333,334,335,336,337,338,339,340,341,342,343],"Natywnie obsługuje tekst, obraz i audio.","Bardzo długi kontekst 1M tokenów.","Open weights i licencja Apache 2.0 sprzyjają użyciu komercyjnemu i fine-tuningowi.","Dobre dopasowanie do agentic coding, tool use, RAG i modeli konwersacyjnych.","Dostępne ścieżki wdrożenia: self-hosting, Tinker i partnerzy API.","Open weights under permissive Apache 2.0 license","Multimodal input (text, images, audio)","Extremely large context window (1M tokens)","Adjustable reasoning effort for cost control","Fine-tuning available via Tinker","Supports tool calling and structured output",[345,346,347,348],"Bardzo duży model, więc self-hosting wymaga znacznych zasobów GPU i infrastruktury.","Model nie jest najsilniejszy ogólnie spośród modeli open i closed, według własnego opisu producenta.","Jak każdy duży model bazowy może halucynować, gorzej trzymać się instrukcji w długich rozmowach i wykazywać bias zależny od danych treningowych.","Producer notes residual risk with role-play and indirectly framed harmful prompts; downstream defenses are recommended.",[350,353,355,357,359,362,365,368,371,374,377,380,382,384,387,389,390,393,394,395,398,400,401,403,405,407,410],{"url":351,"text":352},"https:\u002F\u002Ftwitter.com\u002Fintent\u002Ftweet?url=https%3A%2F%2Fwww.aixploria.com%2Finkling-thinking-machines%2F&text=Inkling%3A+D%C3%A9couvrez+le+premier+LLM+de+chez+Thinking+Machines+%3A+un+mod%C3%A8le+ouvert+et+multimodal+avec+un+contexte+allant+jusqu%27%C3%A0+1+million+de+tokens.+Il+a+%C3%A9t%C3%A9+con%C3%A7u+principalement+pour+le+code%2C+la+manipulation+d%27outils+et+le+raisonnement+adaptatif+%28avec+des+possibilit%C3%A9s+de+fine-tuning+via+Tinker%29.","",{"url":354,"text":352},"https:\u002F\u002Fwww.facebook.com\u002Fsharer\u002Fsharer.php?u=https%3A%2F%2Fwww.aixploria.com%2Finkling-thinking-machines%2F",{"url":356,"text":352},"https:\u002F\u002Fwww.linkedin.com\u002FshareArticle?url=https%3A%2F%2Fwww.aixploria.com%2Finkling-thinking-machines%2F&title=Inkling",{"url":358,"text":352},"https:\u002F\u002Ft.me\u002Fshare\u002Furl?url=https%3A%2F%2Fwww.aixploria.com%2Finkling-thinking-machines%2F&text=D%C3%A9couvrez+le+premier+LLM+de+chez+Thinking+Machines+%3A+un+mod%C3%A8le+ouvert+et+multimodal+avec+un+contexte+allant+jusqu%27%C3%A0+1+million+de+tokens.+Il+a+%C3%A9t%C3%A9+con%C3%A7u+principalement+pour+le+code%2C+la+manipulation+d%27outils+et+le+raisonnement+adaptatif+%28avec+des+possibilit%C3%A9s+de+fine-tuning+via+Tinker%29.",{"url":360,"text":361},"https:\u002F\u002Fchatgpt.com\u002F?q=Donne-moi%20une%20pr%C3%A9sentation%20claire%20et%20concise%20de%20Inkling.%20Explique%20%C3%A0%20quoi%20sert%20cet%20outil%20IA%2C%20%C3%A0%20qui%20il%20s%E2%80%99adresse%2C%20ses%20principales%20fonctionnalit%C3%A9s%2C%20son%20mod%C3%A8le%20tarifaire%2C%20ses%20points%20forts%2C%20ses%20limites%20et%20ses%20alternatives%20possibles.","ChatGPT",{"url":363,"text":364},"https:\u002F\u002Fwww.perplexity.ai\u002F?q=Recherche%20des%20informations%20%C3%A0%20jour%20sur%20Inkling.%20R%C3%A9sume%20%C3%A0%20quoi%20sert%20cet%20outil%2C%20ses%20prix%20actuels%20si%20disponibles%2C%20ses%20principales%20fonctionnalit%C3%A9s%2C%20ses%20points%20forts%2C%20ses%20limites%20et%20ses%20alternatives%20notables.","Perplexity",{"url":366,"text":367},"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002Finkling","Hugging Face",{"url":369,"text":370},"https:\u002F\u002Ftinker.thinkingmachines.ai\u002Fplayground","Playground",{"url":372,"text":373},"https:\u002F\u002Fthinkingmachines.ai\u002Fnews\u002Fintroducing-inkling\u002F","Info",{"url":375,"text":376},"https:\u002F\u002Ffr.wikipedia.org\u002Fwiki\u002FGrand_mod%C3%A8le_de_langage","modèle LLM",{"url":378,"text":379},"https:\u002F\u002Fget.emergent.sh\u002Fe8utricukk6y","Emergent AI",{"url":378,"text":381},"« Transformez vos idées en applications complètes grâce à des agents IA qui codent, testent et déploient pour vous. Créez sites web (full-stack), apps IA, jeux et outils automatisé",{"url":378,"text":383},"Visiter",{"url":385,"text":386},"https:\u002F\u002Ftry.web.clickup.com\u002Fgslcwsp7r8e2","ClickUp",{"url":385,"text":388},"« Gérez vos projets plus vite avec une plateforme qui regroupe équipes, tâches et documents en un seul espace. Son IA intégrée résume vos notes, rédige pour vous et organise automa",{"url":385,"text":383},{"url":391,"text":392},"https:\u002F\u002Fwww.thumbnailcreator.com\u002F","ThumbnailCreator.com",{"url":391,"text":322},{"url":391,"text":383},{"url":396,"text":397},"https:\u002F\u002Fvmake.ai\u002Fvideo-enhancer","Vmake Video Enhancer",{"url":396,"text":399},"« Améliorez vos vidéos floues et de mauvaise qualité pour les transformer en HD et 4K grâce à l'IA. Vmake accentue les détails de vos vidéos, améliore leur netteté et leur donne un",{"url":396,"text":383},{"url":402,"text":352},"https:\u002F\u002Ftwitter.com\u002FAixploria",{"url":404,"text":352},"https:\u002F\u002Fwww.youtube.com\u002F@aixploria-off?sub_confirmation=1",{"url":406,"text":352},"https:\u002F\u002Fwww.tiktok.com\u002F@aixploria",{"url":408,"text":409},"https:\u002F\u002Fcookiedatabase.org\u002Ftcf\u002Fpurposes\u002F","En savoir plus sur ces finalités",{"url":408,"text":409},[],[413,414,415,416,417,418,419,420,421,422],{"url":358,"text":352},{"url":360,"text":361},{"url":363,"text":364},{"url":366,"text":367},{"url":369,"text":370},{"url":372,"text":373},{"url":375,"text":376},{"url":378,"text":379},{"url":378,"text":381},{"url":378,"text":383},"Inkling : Avis, Prix, Info & 41 IA Alternatives | 2026 | AIxploria \nTop 10 Catégories IA Tutoriels IA + Plus Nouveaux Outils IA Outils IA Gratuits Actualités IA 2026 Bonus & Extras Soumettre une IA \nEnglish Français Español Deutsch Italiano Português + More × 中文 日本語 العربية हिन्दी Русский 한국어 Türkçe Tiếng Việt فارسی বাংলা Indonesia ไทย Nederlands Polski اردو Kiswahili Українська Ελληνικά עברית Svenska Norsk Dansk Suomi Čeština Magyar Melayu Tagalog 中文 (香港) தமிழ் \nS'inscrire \nTop 10 Catégories IA Tutoriels IA + Plus Nouveaux Outils IA Outils IA Gratuits Actualités IA 2026 Bonus & Extras Soumettre une IA \nS'inscrire \nAccueil IA Récentes Inkling \nInkling \n#25 dans Modèles LLM \n4.4\u002F5 \nVisiter ce site \n« Découvrez le premier LLM de chez Thinking Machines : un modèle ouvert et multimodal avec un contexte allant jusqu'à 1 million de tokens. Il a été conçu principalement pour le code, la manipulation d'outils et le raisonnement adaptatif (avec des possibilités de fine-tuning via Tinker).\n»\nEn savoir plus : \nChatGPT \nPerplexity \nGratuit \n59924 \nLiens utiles \nSite Officiel \nHugging Face \nPlayground \nInfo \n#IA Récentes \n#Modèles LLM C'est un nouvel outil IA \n69 Inkling : 975 milliards de paramètres à poids ouverts signés Mira Murati\nInkling est le premier modèle d'IA en poids ouverts de Thinking Machines Lab, le laboratoire fondé par Mira Murati, ancienne directrice technique d'OpenAI. Ses poids se téléchargent gratuitement sous licence Apache 2.0; seule l'inférence, sur vos machines ou via une API partenaire, se paie. Sous le capot, un mélange d'experts de 975 milliards de paramètres , dont 41 milliards actifs par requête, et une fenêtre de contexte d'un million de tokens. Il lit texte, images et audio, puis répond en texte, code compris.\nAvantages Poids téléchargeables et modifiables sous licence Apache 2.0 Comprend nativement texte, images et audio Fenêtre de contexte d'un million de tokens Effort de réflexion réglable pour maîtriser les coûts Déclinaison Inkling-Small pour les configurations modestes Inconvénients Sorties limitées au texte, sans génération d'images Version complète réservée aux gros clusters GPU Performance brute en retrait des tout meilleurs modèles Trois sens en entrée, un curseur d'effort en sortie\nInkling accepte du texte, des images et de l'audio, et produit du texte, y compris du code et des données structurées. Ce modèle LLM repose sur une architecture en mélange d'experts: sur ses 975 milliards de paramètres, seuls 41 milliards se réveillent à chaque requête, ce qui allège la facture d'inférence. L'entraînement a englouti 45 000 milliards de tokens mêlant texte, images, audio et vidéo.\nSon réglage le plus malin au quotidien: l'effort de réflexion. Vous baissez le curseur, la réponse tombe plus vite et consomme moins; vous le montez, le modèle creuse. Sur un test de code, Thinking Machines annonce le même score que le Nemotron 3 Ultra de NVIDIA avec trois fois moins de tokens. Et plutôt que d'inventer, Inkling a été entraîné à signaler ce dont il n'est pas sûr.\nBenchmarks, architecture et tests réels d'Inkling Première prise en main d'Inkling, du code au multimodal Inkling, une base à façonner plutôt qu'un trophée de podium\nThinking Machines assume qu'Inkling n'est pas le modèle le plus fort du marché, poids ouverts ou fermés confondus. Le pari est ailleurs: livrer un socle équilibré que chaque organisation affine ensuite sur ses propres données via Tinker, le service maison de fine-tuning, avec des contextes de 64K ou 256K tokens. À sa sortie, il était le plus gros modèle américain à poids ouverts, en réplique directe aux DeepSeek V4, GLM 5.2 et Kimi K2.6 qui trustent le haut du classement des modèles LLM .\nQuelques signes particuliers relevés au lancement:\n77,6 % sur SWE-bench Verified, devant Nemotron 3 en ingénierie logicielle parmi les modèles ouverts les plus solides en compréhension audio (VoiceBench, MMAU) capable d'écrire ses propres scripts de fine-tuning via un agent de code entraîné à répondre franchement sur les sujets exposés à la censure un petit frère, Inkling-Small, 276 milliards de paramètres dont 12 actifs Où essayer et télécharger Inkling\nLe chemin le plus court passe par l'Inkling Playground, un chat mis en ligne gratuitement au lancement. Pour un usage sérieux, plusieurs voies coexistent (votre PC de bureau, lui, déclinera poliment l'invitation pour la version complète).\nVoie d'accès Ce que vous obtenez Pour qui Inkling Playground essai conversationnel dans le navigateur les curieux, sans installation Hugging Face poids complets BF16, NVFP4 et variantes GGUF équipes disposant d'un cluster GPU Tinker fine-tuning sur vos données, contexte 64K ou 256K personnalisation métier API partenaires (Together AI, Fireworks, Baseten, Modal, Databricks) inférence hébergée facturée à l'usage intégration dans vos applications Essayer Inkling gratuitement Questions fréquentes\nInkling est-il gratuit ? Oui, les poids d'Inkling se téléchargent gratuitement sur Hugging Face sous licence Apache 2.0. Le coût se déplace vers l'exécution: soit votre propre matériel, soit une API facturée à l'usage chez Together AI, Fireworks, Baseten, Modal ou Databricks. Le fine-tuning via Tinker se paie également.\nInkling est-il vraiment open source ? Techniquement, Inkling est un modèle à poids ouverts plutôt qu'open source complet: la licence Apache 2.0 couvre usage commercial, modification et redistribution sans redevance, mais les données et le code d'entraînement restent privés. C'est le même standard que les publications de Llama ou DeepSeek.\nQuelle machine faut-il pour faire tourner Inkling en local ? Du très lourd: le point de contrôle BF16 réclame 2 To de VRAM cumulée, la version quantisée NVFP4 se contente d'environ 600 Go, et des GGUF compressés circulent pour llama.cpp. Sur une machine modeste, la piste réaliste s'appelle Inkling-Small et ses 12 milliards de paramètres actifs.\nInkling ou DeepSeek, lequel choisir ? Les deux jouent dans la même cour des grands modèles ouverts, DeepSeek V4 côté chinois, Inkling côté américain. Les premiers comparatifs donnent l'avantage à certains rivaux sur le code en terminal, mais Inkling reste le seul du lot à écouter de l'audio et lire des images nativement, avec un contexte d'un million de tokens.\nVerdict : Champion des classements, non; matière première, oui: si votre équipe cherche un modèle multimodal à affiner sur ses propres données sans dépendre d'une API fermée, Inkling est l'une des fondations ouvertes les plus ambitieuses à l'ouest du Pacifique.\n★ Outils IA de Prestige ★ \n★ Mis en avant \nEmergent AI \n« Transformez vos idées en applications complètes grâce à des agents IA qui codent, testent et déploient pour vous. Créez sites web (full-stack), apps IA, jeux et outils automatisés simplement en décrivant votre projet »\nVisiter \n★ Mis en avant \nClickUp \n« Gérez vos projets plus vite avec une plateforme qui regroupe équipes, tâches et documents en un seul espace. Son IA intégrée résume vos notes, rédige pour vous et organise automatiquement vos priorités »\nVisiter \n★ Mis en avant \nThumbnailCreator.com \n« Créez rapidement des miniatures accrocheuses pour vos vidéos sur YouTube, Twitch et Facebook. Attirez plus de vues et améliorez facilement le taux de clics sur vos clip vidéos »\nVisiter \n★ Mis en avant \nVmake Video Enhancer \n« Améliorez vos vidéos floues et de mauvaise qualité pour les transformer en HD et 4K grâce à l'IA. Vmake accentue les détails de vos vidéos, améliore leur netteté et leur donne un aspect plus professionnel en quelques secondes »\nVisiter \nIA Alternatives à \nInkling \nPayant GPT-5.4 \n4.4\u002F5 \n« Le modèle frontier le plus puissant d'OpenAI contrôle directement votre ordinateur, traite 1 million de tokens et raisonne sur des tâches très longues. Annoncé comme réduisant les erreurs de 33 % et améliorant considérablement le code et l'analyse des documents »\n189 Visiter Gratuit DeepSeek V4 \n4.4\u002F5 \n« Ce modèle Mixture‑of‑Experts d’environ 1 000 milliards de paramètres offre un contexte de 1M tokens, une mémoire Engram quasi infinie et des capacités multimodales texte, image et vidéo. Il vise des scores proches de Claude Opus en code tout en restant bien moins cher ( prévus en Apache 2.0)»\n97 Visiter Freemium Kimi K3 \n4.4\u002F5 \n« Le modèle LLM de chez Moonshot qui rivalise avec les meilleurs modèles propriétaires sur de nombreux benchmarks. Conçu pour le codage longue durée, le raisonnement avancé et les tâches de knowledge work, c'est aussi le premier modèle open source à franchir la barre des 2,8 trillions de paramètres»\n242 Visiter Payant GPT‑5.6 \n4.4\u002F5 \n« Le dernier modèle d’OpenAI pour ceux qui ont besoin d’un IA plus forte en code, en science et en cybersécurité. Disponible directement dans ChatGPT ou via l'API en trois versions : Sol, Terra et Luna»\n102 Visiter Payant Claude Opus 4.8 \n4.4\u002F5 \n« Exploitez le modèle IA d'Anthropic, plus fiable, plus honnête et 4 fois moins enclin à laisser passer ses propres erreurs. Son mode rapide délivre jusqu'à 2,5x plus de tokens par seconde (au même tarif API que la version précédente)\n»\n248 Visiter Freemium GPT‑5.5 Instant \n4.4\u002F5 \n« Utilisez un ChatGPT plus fiable et plus direct. Ce nouveau modèle réduit nettement les hallucinations sur les sujets sensibles (médecine, droit, finance). Il raccourcit ses réponses et personnalise davantage selon vos fichiers, conversations passées et compte Gmail connecté.»\n89 Visiter Freemium Claude Sonnet 5 \n4.5\u002F5 \n« Un modèle LLM pensé pour agir comme un vrai collaborateur technique. Il planifie, utilise le navigateur et le terminal, gère du code sur de longs contextes et atteint des scores de niveau Opus. Il reste également moins cher et plus adapté aux agents autonomes»\n104 Visiter Payant Claude Opus 4.7 \n4.5\u002F5 \n« Le dernier modèle haut de gamme de chez Anthropic optimisé pour le code agentique, les tâches longues et les scénarios critiques en entreprise. Avec auto‑vérification des réponses et fenêtre de contexte d’un million de tokens. Il est également capable d'ajuster automatiquement la profondeur de sa réflexion selon la difficulté du problème»\n162 Visiter \n+ Voir plus d'IA \nCopier le code \nL'intelligence artificielle pour tous\nRessources Tutoriels, astuces et blog Agenda des conférences sur l'IA Glossaire des IA Emplois dans l'IA Newsletter Services pratiques Chaînes YouTube sur l'IA Top 100 IA Liste GPTs Outils IA Hubspot Meilleurs Agents IA Entreprise Ajouter un outil IA Annoncer Mettre à jour votre site Sponsorship ★ Media Kit Qui sommes-nous ? Contactez-nous \n© 2026 AIxploria. Tous droits réservés \nPolitique des Cookies \nConditions d'utilisations Mentions Légales Gérer le consentement aux cookies \nAfin de fournir une expérience optimale, Aixploria utilise des technologies telles que les cookies pour stocker et\u002Fou accéder aux informations relatives à l'appareil. En acceptant ces technologies, vous nous permettez de traiter des données telles que le comportement de navigation ou des identifiants uniques sur ce site. \nFonctionnel \nFonctionnel \nToujours activé \nLe stockage ou l’accès technique est strictement nécessaire dans la finalité d’intérêt légitime de permettre l’utilisation d’un service spécifique explicitement demandé par l’abonné ou l’internaute, ou dans le seul but d’effectuer la transmission d’une communication sur un réseau de communications électroniques. \nPréférences \nPréférences \nLe stockage ou l’accès technique est nécessaire dans la finalité d’intérêt légitime de stocker des préférences qui ne sont pas demandées par l’abonné ou la personne utilisant le service. \nStatistiques \nStatistiques \nLe stockage ou l’accès technique qui est utilisé exclusivement à des fins statistiques. \nLe stockage ou l’accès technique qui est utilisé exclusivement dans des finalités statistiques anonymes. En l’absence d’une assignation à comparaître, d’une conformité volontaire de la part de votre fournisseur d’accès à internet ou d’enregistrements supplémentaires provenant d’une tierce partie, les informations stockées ou extra","2026-08-09T00:14:43.170Z",200,"text\u002Fhtml; charset=UTF-8","2c6190c18f5addd48c5cca6607e605bccb20d776de5a38e0e597d41861e294d3","data\u002Fcatalog\u002Fcache\u002Faixploria\u002Fbafa5f6cf2cf2288f618d82e718d5047b8c787281b3fbadc6d2f41fd66873605.html",true,"human_verified","https:\u002F\u002Fthinkingmachines.ai",{"url":366,"final_url":433,"status":425,"fetched_at":434,"content_hash":435,"title":436,"description":437,"headings":438,"pricing_evidence":458,"privacy_evidence":468,"capability_evidence":473,"text_excerpt":480,"linked_pages":481,"research_links":668,"blog_titles":671,"use_case_evidence":673,"feature_evidence":674},"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling","2026-08-09T01:07:48.463Z","071b65bbe9b71bd6c6fa9ff0eaad27a0cffa487372c816a79082d221fb72f273","thinkingmachines\u002FInkling · Hugging Face","We’re on a journey to advance and democratize artificial intelligence through open source and open science.",[439,440,305,441,442,443,444,445,446,447,448,449,450,451,452,453,454,455,456,305,457],"thinkingmachines \u002F Inkling like 1.7k Follow Thinking Machines Lab 1.65k","Instructions to use thinkingmachines\u002FInkling with libraries, inference providers, notebooks, and local apps. Follow these links to get started.","1. General Information","2. Getting Started","3. Model Properties","Model type","Architecture type","Parameters","Numerics support","Input modalities","Output modalities","4. Training","5. Evaluations","6. Safety","7. Bias, risks and limitations","Model tree for thinkingmachines\u002FInkling","Spaces using thinkingmachines\u002FInkling 17","Collection including thinkingmachines\u002FInkling","Evaluation results",[459,460,453,461,462,463,464,465,466,467],"Hugging Face Models Datasets Spaces Buckets new Docs Enterprise Pricing Website Tasks HuggingChat Collections Languages Organizations Community Blog Posts Daily Papers Hardware Learn Discord Forum GitHub Solutions Team & Enterprise Hugging Face PRO Enterprise Support Inference Providers Inference Endpoints Storage Buckets Log In Sign Up \\\", \\\"assistant\\\": \\\"\u003C|message_model|>\\\", \\\"system\\\": \\\"\u003C|message_system|>\\\", \\\"tool\\\": \\\"\u003C|message_tool|>\\\"} -%}\\n\\n{%- macro emit_thinking_effort() -%}\\n {%- set eff = reasoning_effort if reasoning_effort is defined and reasoning_effort is not none else 0.9 -%}\\n {%- if eff is string -%}\\n {%- set key = eff | trim -%}\\n {%- if key not in effort_map -%}\\n {{- raise_exception(\\\"Unknown reasoning_effort: \\\" ~ eff) -}}\\n {%- endif -%}\\n {%- set num = effort_map[key] -%}\\n {%- else -%}\\n {%- set num = eff | float -%}\\n {%- endif -%}\\n {%- if num \u003C 0.0 or num > 0.99 -%}\\n {{- raise_exception(\\\"reasoning_effort must be in [0.0, 0.99]\\\") -}}\\n {%- endif -%}\\n {{- \\\"\u003C|message_system|>\u003C|content_text|>Thinking effort level: \\\" -}}\\n {%- if num == 0.0 -%}0{%- else -%}{{ num }}{%- endif -%}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n{%- endmacro -%}\\n\\n{%- if tools -%}\\n {%- set tool_state = namespace(specs=[]) -%}\\n {%- for tool in tools -%}\\n {%- set fn = tool.function if tool.function is defined else tool -%}\\n {%- set spec = {\\n \\\"description\\\": (fn.description if fn.description is defined and fn.description else \\\"\\\"),\\n \\\"name\\\": fn.name,\\n \\\"parameters\\\": (fn.parameters if fn.parameters is defined and fn.parameters else {}),\\n \\\"type\\\": (tool.type if tool.type is defined and tool.type else \\\"function\\\"),\\n } -%}\\n {%- set tool_state.specs = tool_state.specs + [spec] -%}\\n {%- endfor -%}\\n {{- \\\"\u003C|message_system|>tool_declare\u003C|content_xml|>\\\" -}}\\n {{- tool_state.specs | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\")) -}}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n{%- endif -%}\\n\\n{%- set state = namespace(effort_emitted=false) -%}\\n{%- for message in messages -%}\\n {%- if message.role not in role_token -%}\\n {{- raise_exception(\\\"Unknown message role: \\\" ~ message.role) -}}\\n {%- endif -%}\\n {%- if not state.effort_emitted and message.role != \\\"system\\\" -%}\\n {{- emit_thinking_effort() -}}\\n {%- set state.effort_emitted = true -%}\\n {%- endif -%}\\n\\n {%- set rtok = role_token[message.role] -%}\\n\\n {%- if message.role == \\\"tool\\\" -%}\\n {%- set tool_name_state = namespace(name=\\\"\\\") -%}\\n {%- if message.name is defined and message.name -%}\\n {%- set tool_name_state.name = message.name -%}\\n {%- elif message.tool_call_id is defined and message.tool_call_id -%}\\n {%- for prev in messages -%}\\n {%- if prev.role == \\\"assistant\\\" and prev.tool_calls -%}\\n {%- for tc in prev.tool_calls -%}\\n {%- if tc.id is defined and tc.id == message.tool_call_id and tc.function.name is defined -%}\\n {%- set tool_name_state.name = tc.function.name -%}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n {{- rtok -}}\\n {%- if tool_name_state.name -%}{{- tool_name_state.name -}}{%- endif -%}\\n {{- \\\"\u003C|content_text|>\\\" -}}\\n {%- if message.content is string -%}{{- message.content -}}{%- endif -%}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n\\n {%- else -%}\\n {%- if message.role == \\\"assistant\\\" and message.reasoning_content is defined and message.reasoning_content -%}\\n {{- \\\"\u003C|message_model|>\u003C|content_thinking|>\\\" ~ message.reasoning_content ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- endif -%}\\n\\n {%- if message.content is string -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ message.content ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif message.content -%}\\n {%- for part in message.content -%}\\n {%- if part is string -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ part ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif part.type is not defined or part.type in (\\\"text\\\", \\\"input_text\\\") -%}\\n {%- set text_part = (part.text if part.text is defined and part.text is string else \\\"\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ text_part ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif part.type in (\\\"image\\\", \\\"input_image\\\", \\\"image_url\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_image|>\u003C|unused_200054|>\u003C|end_message|>\\\" -}}\\n {%- elif part.type in (\\\"audio\\\", \\\"input_audio\\\", \\\"audio_url\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_audio_input|>\u003C|unused_200053|>\u003C|audio_end|>\u003C|end_message|>\\\" -}}\\n {%- else -%}\\n {{- raise_exception(\\\"Unsupported content part type: \\\" ~ part.type) -}}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n\\n {%- if message.role == \\\"assistant\\\" and message.tool_calls -%}\\n {%- for tc in message.tool_calls -%}\\n {%- set fn = tc.function -%}\\n {%- if fn.name is not defined or fn.name is not string -%}\\n {{- raise_exception(\\\"tool call function name must be a string\\\") -}}\\n {%- endif -%}\\n {%- set args = fn.arguments if fn.arguments is defined and fn.arguments else {} -%}\\n {%- if args is string -%}\\n {{- raise_exception(\\\"tool call arguments must be a parsed object, not a JSON string; canonicalize upstream\\\") -}}\\n {%- endif -%}\\n {%- if args is not mapping -%}\\n {{- raise_exception(\\\"tool call arguments must be an object\\\") -}}\\n {%- endif -%}\\n {{- \\\"\u003C|message_model|>\\\" ~ fn.name ~ \\\"\u003C|content_invoke_tool_json|>\\\" -}}\\n {{- '{\\\"name\\\":' ~ (fn.name | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\"))) ~ ',\\\"args\\\":' -}}\\n {{- (args | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\"))) -}}\\n {{- \\\"}\u003C|end_message|>\\\" -}}\\n {%- endfor -%}\\n {%- endif -%}\\n\\n {%- if message.role == \\\"assistant\\\" -%}\\n {{- \\\"\u003C|content_model_end_sampling|>\\\" -}}\\n {%- endif -%}\\n {%- endif -%}\\n{%- endfor -%}\\n\\n{%- if not state.effort_emitted -%}\\n {{- emit_thinking_effort() -}}\\n{%- endif -%}\\n\\n{%- if add_generation_prompt -%}\\n {{- \\\"\u003C|message_model|>\\\" -}}\\n{%- endif -%}\\n\"},\"createdAt\":\"2026-07-14T13:23:14.000Z\",\"discussionsDisabled\":false,\"discussionsSorting\":\"recently-created\",\"downloads\":77799,\"downloadsAllTime\":77799,\"id\":\"thinkingmachines\u002FInkling\",\"isLikedByUser\":false,\"availableInferenceProviders\":[{\"provider\":\"baseten\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002Finkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":true,\"toolCalling\":true},\"isCheapestPricingOutput\":true,\"isFastestThroughput\":true,\"isModelAuthor\":false,\"tokensPerSecond\":158.72110743627016,\"pricingOutput\":4.05},{\"provider\":\"fireworks-ai\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"accounts\u002Ffireworks\u002Fmodels\u002Finkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":false,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":102.38685589674759},{\"provider\":\"together\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002FInkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":true,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":127.45434889475416,\"pricingOutput\":4.05},{\"provider\":\"deepinfra\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002FInkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":false,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":99.40961150436104,\"pricingOutput\":4.05}],\"showHuggingChatEntry\":true,\"inference\":\"warm\",\"lastModified\":\"2026-07-23T17:27:17.000Z\",\"likes\":1703,\"pipeline_tag\":\"image-text-to-text\",\"library_name\":\"transformers\",\"librariesOther\":[],\"trackDownloads\":true,\"model-index\":null,\"evalResults\":[{\"dataset\":{\"id\":\"SWE-bench\u002FSWE-bench_Verified\",\"isBenchmark\":true,\"task_id\":\"swe_bench_%_resolved\"},\"value\":77.6,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fswe-bench_verified.yaml\",\"verified\":false,\"rank\":10,\"label\":\"Swe Bench Resolved\"},{\"dataset\":{\"id\":\"ScaleAI\u002FSWE-bench_Pro\",\"isBenchmark\":true,\"task_id\":\"SWE_Bench_Pro\"},\"value\":54.3,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fswe-bench_pro.yaml\",\"verified\":false,\"rank\":16,\"label\":\"SWE Bench Pro\"},{\"dataset\":{\"id\":\"Idavidrein\u002Fgpqa\",\"isBenchmark\":true,\"task_id\":\"diamond\"},\"value\":87.2,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fgpqa.yaml\",\"verified\":false,\"rank\":19,\"label\":\"Diamond\"},{\"dataset\":{\"id\":\"MathArena\u002Faime_2026\",\"isBenchmark\":true,\"task_id\":\"MathArena\u002Faime_2026\"},\"value\":97.1,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Faime_2026.yaml\",\"verified\":false,\"rank\":1,\"label\":\"MathArena Aime 2026\"},{\"dataset\":{\"id\":\"cais\u002Fhle\",\"isBenchmark\":false,\"task_id\":\"hle\"},\"value\":46,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fhle.yaml\",\"verified\":false,\"rank\":14,\"label\":\"Hle\"},{\"dataset\":{\"id\":\"MMMU\u002FMMMU_Pro\",\"isBenchmark\":true,\"task_id\":\"mmmu_pro_standard_10_options\"},\"value\":73.5,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fmmmu_pro.yaml\",\"verified\":false,\"rank\":3,\"label\":\"Mmmu Pro Standard 10 Options\"}],\"private\":false,\"repoType\":\"model\",\"gated\":false,\"tags\":[\"transformers\",\"safetensors\",\"inkling_mm_model\",\"image-text-to-text\",\"conversational\",\"audio-text-to-text\",\"moe\",\"license:apache-2.0\",\"eval-results\",\"endpoints_compatible\",\"region:us\"],\"tag_objs\":[{\"id\":\"image-text-to-text\",\"label\":\"Image-Text-to-Text\",\"type\":\"pipeline_tag\",\"subType\":\"multimodal\"},{\"id\":\"transformers\",\"label\":\"Transformers\",\"type\":\"library\"},{\"id\":\"safetensors\",\"label\":\"Safetensors\",\"type\":\"library\"},{\"id\":\"inkling_mm_model\",\"label\":\"inkling_mm_model\",\"type\":\"other\",\"clickable\":true},{\"id\":\"conversational\",\"label\":\"conversational\",\"type\":\"other\",\"clickable\":true},{\"id\":\"audio-text-to-text\",\"label\":\"audio-text-to-text\",\"type\":\"other\",\"clickable\":true},{\"id\":\"moe\",\"label\":\"Mixture of Experts\",\"type\":\"other\",\"clickable\":true},{\"id\":\"eval-results\",\"label\":\"Eval Results\",\"type\":\"other\",\"clickable\":true},{\"id\":\"endpoints_compatible\",\"label\":\"Inference Endpoints\",\"type\":\"other\",\"clickable\":true},{\"id\":\"license:apache-2.0\",\"label\":\"apache-2.0\",\"type\":\"license\"},{\"type\":\"region\",\"label\":\"🇺🇸 Region: US\",\"id\":\"region:us\"}],\"transformersInfo\":{\"auto_model\":\"AutoModelForMultimodalLM\",\"pipeline_tag\":\"image-text-to-text\",\"processor\":\"AutoProcessor\"},\"widgetData\":[{\"text\":\"Hi, what can you help me with?\"},{\"text\":\"What is 84 * 3 \u002F 2?\"},{\"text\":\"Tell me an interesting fact about the universe!\"},{\"text\":\"Explain quantum computing in simple terms.\"}],\"safetensors\":{\"parameters\":{\"BF16\":952377607178,\"F32\":16448},\"total\":952377623626,\"sharded\":true,\"totalFileSize\":1904755592540},\"hasBlockedOids\":false,\"region\":\"us\",\"isQuantized\":false},\"discussionsStats\":{\"closed\":17,\"open\":0,\"total\":17},\"query\":{},\"inferenceContextData\":{\"billableEntities\":[],\"entityName2Providers\":{},\"defaultProviders\":[{\"isOriginalProvider\":false,\"name\":\"together\",\"enabled\":true,\"position\":5,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"fireworks-ai\",\"enabled\":true,\"position\":6,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"baseten\",\"enabled\":true,\"position\":13,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"deepinfra\",\"enabled\":true,\"position\":15,\"isReleased\":true,\"accuratePricing\":true}]},\"hasQuantizations\":true,\"copyToBucketNamespaces\":[]}\"> thinkingmachines \u002F Inkling like 1.7k Follow Thinking Machines Lab 1.65k","Quantizations to use this model in llama.cpp , Ollama , LM Studio , or any compatible app. Inkling 1. General Information 2. Getting Started 3. Model Properties Model type Architecture type Parameters Numerics support Input modalities Output modalities 4. Training 5. Evaluations 6. Safety 7. Bias, risks and limitations","Inkling may exhibit general limitations common to foundation models, including hallucination (generating plausible but factually incorrect or unsupported content), occasional failures to follow instructions precisely, and degraded performance in long multi-turn conversations. As with other large-scale models trained on web-derived and synthetic data, Inkling may reflect biases present in its training data, including demographic, cultural, or linguistic biases, and may perform unevenly across languages, dialects, or subject domains that were less represented during training.","Inkling's knowledge is limited to information available as of its training cutoff, and it may not reflect events, developments, or changes that occurred afterward.","Implement additional safeguards — such as content filtering, rate limiting, and monitoring — at the application layer, especially for open deployment contexts where Inkling's built-in mitigations may not be sufficient on their own.","\\\", \\\"assistant\\\": \\\"\u003C|message_model|>\\\", \\\"system\\\": \\\"\u003C|message_system|>\\\", \\\"tool\\\": \\\"\u003C|message_tool|>\\\"} -%}\\n\\n{%- macro emit_thinking_effort() -%}\\n {%- set eff = reasoning_effort if reasoning_effort is defined and reasoning_effort is not none else 0.9 -%}\\n {%- if eff is string -%}\\n {%- set key = eff | trim -%}\\n {%- if key not in effort_map -%}\\n {{- raise_exception(\\\"Unknown reasoning_effort: \\\" ~ eff) -}}\\n {%- endif -%}\\n {%- set num = effort_map[key] -%}\\n {%- else -%}\\n {%- set num = eff | float -%}\\n {%- endif -%}\\n {%- if num \u003C 0.0 or num > 0.99 -%}\\n {{- raise_exception(\\\"reasoning_effort must be in [0.0, 0.99]\\\") -}}\\n {%- endif -%}\\n {{- \\\"\u003C|message_system|>\u003C|content_text|>Thinking effort level: \\\" -}}\\n {%- if num == 0.0 -%}0{%- else -%}{{ num }}{%- endif -%}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n{%- endmacro -%}\\n\\n{%- if tools -%}\\n {%- set tool_state = namespace(specs=[]) -%}\\n {%- for tool in tools -%}\\n {%- set fn = tool.function if tool.function is defined else tool -%}\\n {%- set spec = {\\n \\\"description\\\": (fn.description if fn.description is defined and fn.description else \\\"\\\"),\\n \\\"name\\\": fn.name,\\n \\\"parameters\\\": (fn.parameters if fn.parameters is defined and fn.parameters else {}),\\n \\\"type\\\": (tool.type if tool.type is defined and tool.type else \\\"function\\\"),\\n } -%}\\n {%- set tool_state.specs = tool_state.specs + [spec] -%}\\n {%- endfor -%}\\n {{- \\\"\u003C|message_system|>tool_declare\u003C|content_xml|>\\\" -}}\\n {{- tool_state.specs | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\")) -}}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n{%- endif -%}\\n\\n{%- set state = namespace(effort_emitted=false) -%}\\n{%- for message in messages -%}\\n {%- if message.role not in role_token -%}\\n {{- raise_exception(\\\"Unknown message role: \\\" ~ message.role) -}}\\n {%- endif -%}\\n {%- if not state.effort_emitted and message.role != \\\"system\\\" -%}\\n {{- emit_thinking_effort() -}}\\n {%- set state.effort_emitted = true -%}\\n {%- endif -%}\\n\\n {%- set rtok = role_token[message.role] -%}\\n\\n {%- if message.role == \\\"tool\\\" -%}\\n {%- set tool_name_state = namespace(name=\\\"\\\") -%}\\n {%- if message.name is defined and message.name -%}\\n {%- set tool_name_state.name = message.name -%}\\n {%- elif message.tool_call_id is defined and message.tool_call_id -%}\\n {%- for prev in messages -%}\\n {%- if prev.role == \\\"assistant\\\" and prev.tool_calls -%}\\n {%- for tc in prev.tool_calls -%}\\n {%- if tc.id is defined and tc.id == message.tool_call_id and tc.function.name is defined -%}\\n {%- set tool_name_state.name = tc.function.name -%}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n {{- rtok -}}\\n {%- if tool_name_state.name -%}{{- tool_name_state.name -}}{%- endif -%}\\n {{- \\\"\u003C|content_text|>\\\" -}}\\n {%- if message.content is string -%}{{- message.content -}}{%- endif -%}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n\\n {%- else -%}\\n {%- if message.role == \\\"assistant\\\" and message.reasoning_content is defined and message.reasoning_content -%}\\n {{- \\\"\u003C|message_model|>\u003C|content_thinking|>\\\" ~ message.reasoning_content ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- endif -%}\\n\\n {%- if message.content is string -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ message.content ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif message.content -%}\\n {%- for part in message.content -%}\\n {%- if part is string -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ part ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif part.type is not defined or part.type in (\\\"text\\\", \\\"input_text\\\") -%}\\n {%- set text_part = (part.text if part.text is defined and part.text is string else \\\"\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ text_part ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif part.type in (\\\"image\\\", \\\"input_image\\\", \\\"image_url\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_image|>\u003C|unused_200054|>\u003C|end_message|>\\\" -}}\\n {%- elif part.type in (\\\"audio\\\", \\\"input_audio\\\", \\\"audio_url\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_audio_input|>\u003C|unused_200053|>\u003C|audio_end|>\u003C|end_message|>\\\" -}}\\n {%- else -%}\\n {{- raise_exception(\\\"Unsupported content part type: \\\" ~ part.type) -}}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n\\n {%- if message.role == \\\"assistant\\\" and message.tool_calls -%}\\n {%- for tc in message.tool_calls -%}\\n {%- set fn = tc.function -%}\\n {%- if fn.name is not defined or fn.name is not string -%}\\n {{- raise_exception(\\\"tool call function name must be a string\\\") -}}\\n {%- endif -%}\\n {%- set args = fn.arguments if fn.arguments is defined and fn.arguments else {} -%}\\n {%- if args is string -%}\\n {{- raise_exception(\\\"tool call arguments must be a parsed object, not a JSON string; canonicalize upstream\\\") -}}\\n {%- endif -%}\\n {%- if args is not mapping -%}\\n {{- raise_exception(\\\"tool call arguments must be an object\\\") -}}\\n {%- endif -%}\\n {{- \\\"\u003C|message_model|>\\\" ~ fn.name ~ \\\"\u003C|content_invoke_tool_json|>\\\" -}}\\n {{- '{\\\"name\\\":' ~ (fn.name | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\"))) ~ ',\\\"args\\\":' -}}\\n {{- (args | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\"))) -}}\\n {{- \\\"}\u003C|end_message|>\\\" -}}\\n {%- endfor -%}\\n {%- endif -%}\\n\\n {%- if message.role == \\\"assistant\\\" -%}\\n {{- \\\"\u003C|content_model_end_sampling|>\\\" -}}\\n {%- endif -%}\\n {%- endif -%}\\n{%- endfor -%}\\n\\n{%- if not state.effort_emitted -%}\\n {{- emit_thinking_effort() -}}\\n{%- endif -%}\\n\\n{%- if add_generation_prompt -%}\\n {{- \\\"\u003C|message_model|>\\\" -}}\\n{%- endif -%}\\n\"},\"createdAt\":\"2026-07-14T13:23:14.000Z\",\"discussionsDisabled\":false,\"discussionsSorting\":\"recently-created\",\"downloads\":77799,\"downloadsAllTime\":77799,\"id\":\"thinkingmachines\u002FInkling\",\"isLikedByUser\":false,\"availableInferenceProviders\":[{\"provider\":\"baseten\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002Finkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":true,\"toolCalling\":true},\"isCheapestPricingOutput\":true,\"isFastestThroughput\":true,\"isModelAuthor\":false,\"tokensPerSecond\":158.72110743627016,\"pricingOutput\":4.05},{\"provider\":\"fireworks-ai\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"accounts\u002Ffireworks\u002Fmodels\u002Finkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":false,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":102.38685589674759},{\"provider\":\"together\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002FInkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":true,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":127.45434889475416,\"pricingOutput\":4.05},{\"provider\":\"deepinfra\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002FInkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":false,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":99.40961150436104,\"pricingOutput\":4.05}],\"showHuggingChatEntry\":true,\"inference\":\"warm\",\"lastModified\":\"2026-07-23T17:27:17.000Z\",\"likes\":1703,\"pipeline_tag\":\"image-text-to-text\",\"library_name\":\"transformers\",\"librariesOther\":[],\"trackDownloads\":true,\"model-index\":null,\"evalResults\":[{\"dataset\":{\"id\":\"SWE-bench\u002FSWE-bench_Verified\",\"isBenchmark\":true,\"task_id\":\"swe_bench_%_resolved\"},\"value\":77.6,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fswe-bench_verified.yaml\",\"verified\":false,\"rank\":10,\"label\":\"Swe Bench Resolved\"},{\"dataset\":{\"id\":\"ScaleAI\u002FSWE-bench_Pro\",\"isBenchmark\":true,\"task_id\":\"SWE_Bench_Pro\"},\"value\":54.3,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fswe-bench_pro.yaml\",\"verified\":false,\"rank\":16,\"label\":\"SWE Bench Pro\"},{\"dataset\":{\"id\":\"Idavidrein\u002Fgpqa\",\"isBenchmark\":true,\"task_id\":\"diamond\"},\"value\":87.2,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fgpqa.yaml\",\"verified\":false,\"rank\":19,\"label\":\"Diamond\"},{\"dataset\":{\"id\":\"MathArena\u002Faime_2026\",\"isBenchmark\":true,\"task_id\":\"MathArena\u002Faime_2026\"},\"value\":97.1,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Faime_2026.yaml\",\"verified\":false,\"rank\":1,\"label\":\"MathArena Aime 2026\"},{\"dataset\":{\"id\":\"cais\u002Fhle\",\"isBenchmark\":false,\"task_id\":\"hle\"},\"value\":46,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fhle.yaml\",\"verified\":false,\"rank\":14,\"label\":\"Hle\"},{\"dataset\":{\"id\":\"MMMU\u002FMMMU_Pro\",\"isBenchmark\":true,\"task_id\":\"mmmu_pro_standard_10_options\"},\"value\":73.5,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fmmmu_pro.yaml\",\"verified\":false,\"rank\":3,\"label\":\"Mmmu Pro Standard 10 Options\"}],\"private\":false,\"repoType\":\"model\",\"gated\":false,\"tags\":[\"transformers\",\"safetensors\",\"inkling_mm_model\",\"image-text-to-text\",\"conversational\",\"audio-text-to-text\",\"moe\",\"license:apache-2.0\",\"eval-results\",\"endpoints_compatible\",\"region:us\"],\"tag_objs\":[{\"id\":\"image-text-to-text\",\"label\":\"Image-Text-to-Text\",\"type\":\"pipeline_tag\",\"subType\":\"multimodal\"},{\"id\":\"transformers\",\"label\":\"Transformers\",\"type\":\"library\"},{\"id\":\"safetensors\",\"label\":\"Safetensors\",\"type\":\"library\"},{\"id\":\"inkling_mm_model\",\"label\":\"inkling_mm_model\",\"type\":\"other\",\"clickable\":true},{\"id\":\"conversational\",\"label\":\"conversational\",\"type\":\"other\",\"clickable\":true},{\"id\":\"audio-text-to-text\",\"label\":\"audio-text-to-text\",\"type\":\"other\",\"clickable\":true},{\"id\":\"moe\",\"label\":\"Mixture of Experts\",\"type\":\"other\",\"clickable\":true},{\"id\":\"eval-results\",\"label\":\"Eval Results\",\"type\":\"other\",\"clickable\":true},{\"id\":\"endpoints_compatible\",\"label\":\"Inference Endpoints\",\"type\":\"other\",\"clickable\":true},{\"id\":\"license:apache-2.0\",\"label\":\"apache-2.0\",\"type\":\"license\"},{\"type\":\"region\",\"label\":\"🇺🇸 Region: US\",\"id\":\"region:us\"}],\"transformersInfo\":{\"auto_model\":\"AutoModelForMultimodalLM\",\"pipeline_tag\":\"image-text-to-text\",\"processor\":\"AutoProcessor\"},\"widgetData\":[{\"text\":\"Hi, what can you help me with?\"},{\"text\":\"What is 84 * 3 \u002F 2?\"},{\"text\":\"Tell me an interesting fact about the universe!\"},{\"text\":\"Explain quantum computing in simple terms.\"}],\"safetensors\":{\"parameters\":{\"BF16\":952377607178,\"F32\":16448},\"total\":952377623626,\"sharded\":true,\"totalFileSize\":1904755592540},\"hasBlockedOids\":false,\"region\":\"us\",\"isQuantized\":false},\"inferenceContextData\":{\"billableEntities\":[],\"entityName2Providers\":{},\"defaultProviders\":[{\"isOriginalProvider\":false,\"name\":\"together\",\"enabled\":true,\"position\":5,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"fireworks-ai\",\"enabled\":true,\"position\":6,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"baseten\",\"enabled\":true,\"position\":13,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"deepinfra\",\"enabled\":true,\"position\":15,\"isReleased\":true,\"accuratePricing\":true}]},\"canWrite\":false}\"> Safetensors Model size 952B params Tensor type BF16 · F32 · Chat template","\\\", \\\"assistant\\\": \\\"\u003C|message_model|>\\\", \\\"system\\\": \\\"\u003C|message_system|>\\\", \\\"tool\\\": \\\"\u003C|message_tool|>\\\"} -%}\\n\\n{%- macro emit_thinking_effort() -%}\\n {%- set eff = reasoning_effort if reasoning_effort is defined and reasoning_effort is not none else 0.9 -%}\\n {%- if eff is string -%}\\n {%- set key = eff | trim -%}\\n {%- if key not in effort_map -%}\\n {{- raise_exception(\\\"Unknown reasoning_effort: \\\" ~ eff) -}}\\n {%- endif -%}\\n {%- set num = effort_map[key] -%}\\n {%- else -%}\\n {%- set num = eff | float -%}\\n {%- endif -%}\\n {%- if num \u003C 0.0 or num > 0.99 -%}\\n {{- raise_exception(\\\"reasoning_effort must be in [0.0, 0.99]\\\") -}}\\n {%- endif -%}\\n {{- \\\"\u003C|message_system|>\u003C|content_text|>Thinking effort level: \\\" -}}\\n {%- if num == 0.0 -%}0{%- else -%}{{ num }}{%- endif -%}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n{%- endmacro -%}\\n\\n{%- if tools -%}\\n {%- set tool_state = namespace(specs=[]) -%}\\n {%- for tool in tools -%}\\n {%- set fn = tool.function if tool.function is defined else tool -%}\\n {%- set spec = {\\n \\\"description\\\": (fn.description if fn.description is defined and fn.description else \\\"\\\"),\\n \\\"name\\\": fn.name,\\n \\\"parameters\\\": (fn.parameters if fn.parameters is defined and fn.parameters else {}),\\n \\\"type\\\": (tool.type if tool.type is defined and tool.type else \\\"function\\\"),\\n } -%}\\n {%- set tool_state.specs = tool_state.specs + [spec] -%}\\n {%- endfor -%}\\n {{- \\\"\u003C|message_system|>tool_declare\u003C|content_xml|>\\\" -}}\\n {{- tool_state.specs | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\")) -}}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n{%- endif -%}\\n\\n{%- set state = namespace(effort_emitted=false) -%}\\n{%- for message in messages -%}\\n {%- if message.role not in role_token -%}\\n {{- raise_exception(\\\"Unknown message role: \\\" ~ message.role) -}}\\n {%- endif -%}\\n {%- if not state.effort_emitted and message.role != \\\"system\\\" -%}\\n {{- emit_thinking_effort() -}}\\n {%- set state.effort_emitted = true -%}\\n {%- endif -%}\\n\\n {%- set rtok = role_token[message.role] -%}\\n\\n {%- if message.role == \\\"tool\\\" -%}\\n {%- set tool_name_state = namespace(name=\\\"\\\") -%}\\n {%- if message.name is defined and message.name -%}\\n {%- set tool_name_state.name = message.name -%}\\n {%- elif message.tool_call_id is defined and message.tool_call_id -%}\\n {%- for prev in messages -%}\\n {%- if prev.role == \\\"assistant\\\" and prev.tool_calls -%}\\n {%- for tc in prev.tool_calls -%}\\n {%- if tc.id is defined and tc.id == message.tool_call_id and tc.function.name is defined -%}\\n {%- set tool_name_state.name = tc.function.name -%}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n {{- rtok -}}\\n {%- if tool_name_state.name -%}{{- tool_name_state.name -}}{%- endif -%}\\n {{- \\\"\u003C|content_text|>\\\" -}}\\n {%- if message.content is string -%}{{- message.content -}}{%- endif -%}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n\\n {%- else -%}\\n {%- if message.role == \\\"assistant\\\" and message.reasoning_content is defined and message.reasoning_content -%}\\n {{- \\\"\u003C|message_model|>\u003C|content_thinking|>\\\" ~ message.reasoning_content ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- endif -%}\\n\\n {%- if message.content is string -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ message.content ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif message.content -%}\\n {%- for part in message.content -%}\\n {%- if part is string -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ part ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif part.type is not defined or part.type in (\\\"text\\\", \\\"input_text\\\") -%}\\n {%- set text_part = (part.text if part.text is defined and part.text is string else \\\"\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ text_part ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif part.type in (\\\"image\\\", \\\"input_image\\\", \\\"image_url\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_image|>\u003C|unused_200054|>\u003C|end_message|>\\\" -}}\\n {%- elif part.type in (\\\"audio\\\", \\\"input_audio\\\", \\\"audio_url\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_audio_input|>\u003C|unused_200053|>\u003C|audio_end|>\u003C|end_message|>\\\" -}}\\n {%- else -%}\\n {{- raise_exception(\\\"Unsupported content part type: \\\" ~ part.type) -}}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n\\n {%- if message.role == \\\"assistant\\\" and message.tool_calls -%}\\n {%- for tc in message.tool_calls -%}\\n {%- set fn = tc.function -%}\\n {%- if fn.name is not defined or fn.name is not string -%}\\n {{- raise_exception(\\\"tool call function name must be a string\\\") -}}\\n {%- endif -%}\\n {%- set args = fn.arguments if fn.arguments is defined and fn.arguments else {} -%}\\n {%- if args is string -%}\\n {{- raise_exception(\\\"tool call arguments must be a parsed object, not a JSON string; canonicalize upstream\\\") -}}\\n {%- endif -%}\\n {%- if args is not mapping -%}\\n {{- raise_exception(\\\"tool call arguments must be an object\\\") -}}\\n {%- endif -%}\\n {{- \\\"\u003C|message_model|>\\\" ~ fn.name ~ \\\"\u003C|content_invoke_tool_json|>\\\" -}}\\n {{- '{\\\"name\\\":' ~ (fn.name | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\"))) ~ ',\\\"args\\\":' -}}\\n {{- (args | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\"))) -}}\\n {{- \\\"}\u003C|end_message|>\\\" -}}\\n {%- endfor -%}\\n {%- endif -%}\\n\\n {%- if message.role == \\\"assistant\\\" -%}\\n {{- \\\"\u003C|content_model_end_sampling|>\\\" -}}\\n {%- endif -%}\\n {%- endif -%}\\n{%- endfor -%}\\n\\n{%- if not state.effort_emitted -%}\\n {{- emit_thinking_effort() -}}\\n{%- endif -%}\\n\\n{%- if add_generation_prompt -%}\\n {{- \\\"\u003C|message_model|>\\\" -}}\\n{%- endif -%}\\n\"},\"createdAt\":\"2026-07-14T13:23:14.000Z\",\"discussionsDisabled\":false,\"discussionsSorting\":\"recently-created\",\"downloads\":77799,\"downloadsAllTime\":77799,\"id\":\"thinkingmachines\u002FInkling\",\"isLikedByUser\":false,\"availableInferenceProviders\":[{\"provider\":\"baseten\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002Finkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":true,\"toolCalling\":true},\"isCheapestPricingOutput\":true,\"isFastestThroughput\":true,\"isModelAuthor\":false,\"tokensPerSecond\":158.72110743627016,\"pricingOutput\":4.05},{\"provider\":\"fireworks-ai\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"accounts\u002Ffireworks\u002Fmodels\u002Finkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":false,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":102.38685589674759},{\"provider\":\"together\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002FInkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":true,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":127.45434889475416,\"pricingOutput\":4.05},{\"provider\":\"deepinfra\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002FInkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":false,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":99.40961150436104,\"pricingOutput\":4.05}],\"showHuggingChatEntry\":true,\"inference\":\"warm\",\"lastModified\":\"2026-07-23T17:27:17.000Z\",\"likes\":1703,\"pipeline_tag\":\"image-text-to-text\",\"library_name\":\"transformers\",\"librariesOther\":[],\"trackDownloads\":true,\"model-index\":null,\"evalResults\":[{\"dataset\":{\"id\":\"SWE-bench\u002FSWE-bench_Verified\",\"isBenchmark\":true,\"task_id\":\"swe_bench_%_resolved\"},\"value\":77.6,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fswe-bench_verified.yaml\",\"verified\":false,\"rank\":10,\"label\":\"Swe Bench Resolved\"},{\"dataset\":{\"id\":\"ScaleAI\u002FSWE-bench_Pro\",\"isBenchmark\":true,\"task_id\":\"SWE_Bench_Pro\"},\"value\":54.3,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fswe-bench_pro.yaml\",\"verified\":false,\"rank\":16,\"label\":\"SWE Bench Pro\"},{\"dataset\":{\"id\":\"Idavidrein\u002Fgpqa\",\"isBenchmark\":true,\"task_id\":\"diamond\"},\"value\":87.2,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fgpqa.yaml\",\"verified\":false,\"rank\":19,\"label\":\"Diamond\"},{\"dataset\":{\"id\":\"MathArena\u002Faime_2026\",\"isBenchmark\":true,\"task_id\":\"MathArena\u002Faime_2026\"},\"value\":97.1,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Faime_2026.yaml\",\"verified\":false,\"rank\":1,\"label\":\"MathArena Aime 2026\"},{\"dataset\":{\"id\":\"cais\u002Fhle\",\"isBenchmark\":false,\"task_id\":\"hle\"},\"value\":46,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fhle.yaml\",\"verified\":false,\"rank\":14,\"label\":\"Hle\"},{\"dataset\":{\"id\":\"MMMU\u002FMMMU_Pro\",\"isBenchmark\":true,\"task_id\":\"mmmu_pro_standard_10_options\"},\"value\":73.5,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fmmmu_pro.yaml\",\"verified\":false,\"rank\":3,\"label\":\"Mmmu Pro Standard 10 Options\"}],\"private\":false,\"repoType\":\"model\",\"gated\":false,\"tags\":[\"transformers\",\"safetensors\",\"inkling_mm_model\",\"image-text-to-text\",\"conversational\",\"audio-text-to-text\",\"moe\",\"license:apache-2.0\",\"eval-results\",\"endpoints_compatible\",\"region:us\"],\"tag_objs\":[{\"id\":\"image-text-to-text\",\"label\":\"Image-Text-to-Text\",\"type\":\"pipeline_tag\",\"subType\":\"multimodal\"},{\"id\":\"transformers\",\"label\":\"Transformers\",\"type\":\"library\"},{\"id\":\"safetensors\",\"label\":\"Safetensors\",\"type\":\"library\"},{\"id\":\"inkling_mm_model\",\"label\":\"inkling_mm_model\",\"type\":\"other\",\"clickable\":true},{\"id\":\"conversational\",\"label\":\"conversational\",\"type\":\"other\",\"clickable\":true},{\"id\":\"audio-text-to-text\",\"label\":\"audio-text-to-text\",\"type\":\"other\",\"clickable\":true},{\"id\":\"moe\",\"label\":\"Mixture of Experts\",\"type\":\"other\",\"clickable\":true},{\"id\":\"eval-results\",\"label\":\"Eval Results\",\"type\":\"other\",\"clickable\":true},{\"id\":\"endpoints_compatible\",\"label\":\"Inference Endpoints\",\"type\":\"other\",\"clickable\":true},{\"id\":\"license:apache-2.0\",\"label\":\"apache-2.0\",\"type\":\"license\"},{\"type\":\"region\",\"label\":\"🇺🇸 Region: US\",\"id\":\"region:us\"}],\"transformersInfo\":{\"auto_model\":\"AutoModelForMultimodalLM\",\"pipeline_tag\":\"image-text-to-text\",\"processor\":\"AutoProcessor\"},\"widgetData\":[{\"text\":\"Hi, what can you help me with?\"},{\"text\":\"What is 84 * 3 \u002F 2?\"},{\"text\":\"Tell me an interesting fact about the universe!\"},{\"text\":\"Explain quantum computing in simple terms.\"}],\"safetensors\":{\"parameters\":{\"BF16\":952377607178,\"F32\":16448},\"total\":952377623626,\"sharded\":true,\"totalFileSize\":1904755592540},\"hasBlockedOids\":false,\"region\":\"us\",\"isQuantized\":false},\"queryParams\":{},\"inferenceContextData\":{\"billableEntities\":[],\"entityName2Providers\":{},\"defaultProviders\":[{\"isOriginalProvider\":false,\"name\":\"together\",\"enabled\":true,\"position\":5,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"fireworks-ai\",\"enabled\":true,\"position\":6,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"baseten\",\"enabled\":true,\"position\":13,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"deepinfra\",\"enabled\":true,\"position\":15,\"isReleased\":true,\"accuratePricing\":true}]},\"linkToInferenceDiscussion\":{\"url\":\"\u002Fspaces\u002Fhuggingface\u002FInferenceSupport\u002Fdiscussions\u002F11084\",\"numReactions\":56},\"isLoggedIn\":false,\"canWrite\":false,\"examples\":[{\"text\":\"Hi, what can you help me with?\"},{\"text\":\"What is 84 * 3 \u002F 2?\"},{\"text\":\"Tell me an interesting fact about the universe!\"},{\"text\":\"Explain quantum computing in simple terms.\"}],\"initialSelectedProvider\":\"together\",\"widgetType\":\"conversational\",\"isMaximized\":false,\"isMac\":false}}\"> Inference Providers NEW Together AI +1 Image-Text-to-Text Examples Input a message to start chatting with thinkingmachines\u002FInkling . Send View Code Snippets Compare providers Model tree for thinkingmachines\u002FInkling","Collection Inkling is a versatile, customizable model that reasons over text, images, audio, with variable and efficient thinking effort. • 4 items • Updated 12 days ago • 47 \\\", \\\"assistant\\\": \\\"\u003C|message_model|>\\\", \\\"system\\\": \\\"\u003C|message_system|>\\\", \\\"tool\\\": \\\"\u003C|message_tool|>\\\"} -%}\\n\\n{%- macro emit_thinking_effort() -%}\\n {%- set eff = reasoning_effort if reasoning_effort is defined and reasoning_effort is not none else 0.9 -%}\\n {%- if eff is string -%}\\n {%- set key = eff | trim -%}\\n {%- if key not in effort_map -%}\\n {{- raise_exception(\\\"Unknown reasoning_effort: \\\" ~ eff) -}}\\n {%- endif -%}\\n {%- set num = effort_map[key] -%}\\n {%- else -%}\\n {%- set num = eff | float -%}\\n {%- endif -%}\\n {%- if num \u003C 0.0 or num > 0.99 -%}\\n {{- raise_exception(\\\"reasoning_effort must be in [0.0, 0.99]\\\") -}}\\n {%- endif -%}\\n {{- \\\"\u003C|message_system|>\u003C|content_text|>Thinking effort level: \\\" -}}\\n {%- if num == 0.0 -%}0{%- else -%}{{ num }}{%- endif -%}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n{%- endmacro -%}\\n\\n{%- if tools -%}\\n {%- set tool_state = namespace(specs=[]) -%}\\n {%- for tool in tools -%}\\n {%- set fn = tool.function if tool.function is defined else tool -%}\\n {%- set spec = {\\n \\\"description\\\": (fn.description if fn.description is defined and fn.description else \\\"\\\"),\\n \\\"name\\\": fn.name,\\n \\\"parameters\\\": (fn.parameters if fn.parameters is defined and fn.parameters else {}),\\n \\\"type\\\": (tool.type if tool.type is defined and tool.type else \\\"function\\\"),\\n } -%}\\n {%- set tool_state.specs = tool_state.specs + [spec] -%}\\n {%- endfor -%}\\n {{- \\\"\u003C|message_system|>tool_declare\u003C|content_xml|>\\\" -}}\\n {{- tool_state.specs | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\")) -}}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n{%- endif -%}\\n\\n{%- set state = namespace(effort_emitted=false) -%}\\n{%- for message in messages -%}\\n {%- if message.role not in role_token -%}\\n {{- raise_exception(\\\"Unknown message role: \\\" ~ message.role) -}}\\n {%- endif -%}\\n {%- if not state.effort_emitted and message.role != \\\"system\\\" -%}\\n {{- emit_thinking_effort() -}}\\n {%- set state.effort_emitted = true -%}\\n {%- endif -%}\\n\\n {%- set rtok = role_token[message.role] -%}\\n\\n {%- if message.role == \\\"tool\\\" -%}\\n {%- set tool_name_state = namespace(name=\\\"\\\") -%}\\n {%- if message.name is defined and message.name -%}\\n {%- set tool_name_state.name = message.name -%}\\n {%- elif message.tool_call_id is defined and message.tool_call_id -%}\\n {%- for prev in messages -%}\\n {%- if prev.role == \\\"assistant\\\" and prev.tool_calls -%}\\n {%- for tc in prev.tool_calls -%}\\n {%- if tc.id is defined and tc.id == message.tool_call_id and tc.function.name is defined -%}\\n {%- set tool_name_state.name = tc.function.name -%}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n {{- rtok -}}\\n {%- if tool_name_state.name -%}{{- tool_name_state.name -}}{%- endif -%}\\n {{- \\\"\u003C|content_text|>\\\" -}}\\n {%- if message.content is string -%}{{- message.content -}}{%- endif -%}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n\\n {%- else -%}\\n {%- if message.role == \\\"assistant\\\" and message.reasoning_content is defined and message.reasoning_content -%}\\n {{- \\\"\u003C|message_model|>\u003C|content_thinking|>\\\" ~ message.reasoning_content ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- endif -%}\\n\\n {%- if message.content is string -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ message.content ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif message.content -%}\\n {%- for part in message.content -%}\\n {%- if part is string -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ part ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif part.type is not defined or part.type in (\\\"text\\\", \\\"input_text\\\") -%}\\n {%- set text_part = (part.text if part.text is defined and part.text is string else \\\"\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ text_part ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif part.type in (\\\"image\\\", \\\"input_image\\\", \\\"image_url\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_image|>\u003C|unused_200054|>\u003C|end_message|>\\\" -}}\\n {%- elif part.type in (\\\"audio\\\", \\\"input_audio\\\", \\\"audio_url\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_audio_input|>\u003C|unused_200053|>\u003C|audio_end|>\u003C|end_message|>\\\" -}}\\n {%- else -%}\\n {{- raise_exception(\\\"Unsupported content part type: \\\" ~ part.type) -}}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n\\n {%- if message.role == \\\"assistant\\\" and message.tool_calls -%}\\n {%- for tc in message.tool_calls -%}\\n {%- set fn = tc.function -%}\\n {%- if fn.name is not defined or fn.name is not string -%}\\n {{- raise_exception(\\\"tool call function name must be a string\\\") -}}\\n {%- endif -%}\\n {%- set args = fn.arguments if fn.arguments is defined and fn.arguments else {} -%}\\n {%- if args is string -%}\\n {{- raise_exception(\\\"tool call arguments must be a parsed object, not a JSON string; canonicalize upstream\\\") -}}\\n {%- endif -%}\\n {%- if args is not mapping -%}\\n {{- raise_exception(\\\"tool call arguments must be an object\\\") -}}\\n {%- endif -%}\\n {{- \\\"\u003C|message_model|>\\\" ~ fn.name ~ \\\"\u003C|content_invoke_tool_json|>\\\" -}}\\n {{- '{\\\"name\\\":' ~ (fn.name | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\"))) ~ ',\\\"args\\\":' -}}\\n {{- (args | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\"))) -}}\\n {{- \\\"}\u003C|end_message|>\\\" -}}\\n {%- endfor -%}\\n {%- endif -%}\\n\\n {%- if message.role == \\\"assistant\\\" -%}\\n {{- \\\"\u003C|content_model_end_sampling|>\\\" -}}\\n {%- endif -%}\\n {%- endif -%}\\n{%- endfor -%}\\n\\n{%- if not state.effort_emitted -%}\\n {{- emit_thinking_effort() -}}\\n{%- endif -%}\\n\\n{%- if add_generation_prompt -%}\\n {{- \\\"\u003C|message_model|>\\\" -}}\\n{%- endif -%}\\n\"},\"createdAt\":\"2026-07-14T13:23:14.000Z\",\"discussionsDisabled\":false,\"discussionsSorting\":\"recently-created\",\"downloads\":77799,\"downloadsAllTime\":77799,\"id\":\"thinkingmachines\u002FInkling\",\"isLikedByUser\":false,\"availableInferenceProviders\":[{\"provider\":\"baseten\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002Finkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":true,\"toolCalling\":true},\"isCheapestPricingOutput\":true,\"isFastestThroughput\":true,\"isModelAuthor\":false,\"tokensPerSecond\":158.72110743627016,\"pricingOutput\":4.05},{\"provider\":\"fireworks-ai\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"accounts\u002Ffireworks\u002Fmodels\u002Finkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":false,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":102.38685589674759},{\"provider\":\"together\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002FInkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":true,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":127.45434889475416,\"pricingOutput\":4.05},{\"provider\":\"deepinfra\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002FInkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":false,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":99.40961150436104,\"pricingOutput\":4.05}],\"showHuggingChatEntry\":true,\"inference\":\"warm\",\"lastModified\":\"2026-07-23T17:27:17.000Z\",\"likes\":1703,\"pipeline_tag\":\"image-text-to-text\",\"library_name\":\"transformers\",\"librariesOther\":[],\"trackDownloads\":true,\"model-index\":null,\"evalResults\":[{\"dataset\":{\"id\":\"SWE-bench\u002FSWE-bench_Verified\",\"isBenchmark\":true,\"task_id\":\"swe_bench_%_resolved\"},\"value\":77.6,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fswe-bench_verified.yaml\",\"verified\":false,\"rank\":10,\"label\":\"Swe Bench Resolved\"},{\"dataset\":{\"id\":\"ScaleAI\u002FSWE-bench_Pro\",\"isBenchmark\":true,\"task_id\":\"SWE_Bench_Pro\"},\"value\":54.3,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fswe-bench_pro.yaml\",\"verified\":false,\"rank\":16,\"label\":\"SWE Bench Pro\"},{\"dataset\":{\"id\":\"Idavidrein\u002Fgpqa\",\"isBenchmark\":true,\"task_id\":\"diamond\"},\"value\":87.2,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fgpqa.yaml\",\"verified\":false,\"rank\":19,\"label\":\"Diamond\"},{\"dataset\":{\"id\":\"MathArena\u002Faime_2026\",\"isBenchmark\":true,\"task_id\":\"MathArena\u002Faime_2026\"},\"value\":97.1,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Faime_2026.yaml\",\"verified\":false,\"rank\":1,\"label\":\"MathArena Aime 2026\"},{\"dataset\":{\"id\":\"cais\u002Fhle\",\"isBenchmark\":false,\"task_id\":\"hle\"},\"value\":46,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fhle.yaml\",\"verified\":false,\"rank\":14,\"label\":\"Hle\"},{\"dataset\":{\"id\":\"MMMU\u002FMMMU_Pro\",\"isBenchmark\":true,\"task_id\":\"mmmu_pro_standard_10_options\"},\"value\":73.5,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fmmmu_pro.yaml\",\"verified\":false,\"rank\":3,\"label\":\"Mmmu Pro Standard 10 Options\"}],\"private\":false,\"repoType\":\"model\",\"gated\":false,\"tags\":[\"transformers\",\"safetensors\",\"inkling_mm_model\",\"image-text-to-text\",\"conversational\",\"audio-text-to-text\",\"moe\",\"license:apache-2.0\",\"eval-results\",\"endpoints_compatible\",\"region:us\"],\"tag_objs\":[{\"id\":\"image-text-to-text\",\"label\":\"Image-Text-to-Text\",\"type\":\"pipeline_tag\",\"subType\":\"multimodal\"},{\"id\":\"transformers\",\"label\":\"Transformers\",\"type\":\"library\"},{\"id\":\"safetensors\",\"label\":\"Safetensors\",\"type\":\"library\"},{\"id\":\"inkling_mm_model\",\"label\":\"inkling_mm_model\",\"type\":\"other\",\"clickable\":true},{\"id\":\"conversational\",\"label\":\"conversational\",\"type\":\"other\",\"clickable\":true},{\"id\":\"audio-text-to-text\",\"label\":\"audio-text-to-text\",\"type\":\"other\",\"clickable\":true},{\"id\":\"moe\",\"label\":\"Mixture of Experts\",\"type\":\"other\",\"clickable\":true},{\"id\":\"eval-results\",\"label\":\"Eval Results\",\"type\":\"other\",\"clickable\":true},{\"id\":\"endpoints_compatible\",\"label\":\"Inference Endpoints\",\"type\":\"other\",\"clickable\":true},{\"id\":\"license:apache-2.0\",\"label\":\"apache-2.0\",\"type\":\"license\"},{\"type\":\"region\",\"label\":\"🇺🇸 Region: US\",\"id\":\"region:us\"}],\"transformersInfo\":{\"auto_model\":\"AutoModelForMultimodalLM\",\"pipeline_tag\":\"image-text-to-text\",\"processor\":\"AutoProcessor\"},\"widgetData\":[{\"text\":\"Hi, what can you help me with?\"},{\"text\":\"What is 84 * 3 \u002F 2?\"},{\"text\":\"Tell me an interesting fact about the universe!\"},{\"text\":\"Explain quantum computing in simple terms.\"}],\"safetensors\":{\"parameters\":{\"BF16\":952377607178,\"F32\":16448},\"total\":952377623626,\"sharded\":true,\"totalFileSize\":1904755592540},\"hasBlockedOids\":false,\"region\":\"us\",\"isQuantized\":false}}\"> Evaluation results","SWE-bench\u002FSWE-bench_Verified · Swe Bench Resolved View evaluation results leaderboard 77.6 ScaleAI\u002FSWE-bench_Pro · SWE Bench Pro View evaluation results leaderboard 54.3 Idavidrein\u002Fgpqa · Diamond View evaluation results leaderboard 87.2 MathArena\u002Faime_2026 · MathArena Aime 2026 View evaluation results leaderboard 97.1 cais\u002Fhle · Hle View evaluation results 46 MMMU\u002FMMMU_Pro · Mmmu Pro Standard 10 Options View evaluation results leaderboard 73.5 System theme Company TOS Privacy About Careers Website Models Datasets Spaces Pricing Docs",[459,469,470,460,450,471,472,461,462,464,465,466,467],"processor = AutoProcessor.from_pretrained(\"thinkingmachines\u002FInkling\")","model = AutoModelForMultimodalLM.from_pretrained(\"thinkingmachines\u002FInkling\", device_map=\"auto\")","Training data includes a broad variety of content types, including text, images, audio, video. Training data for the model was drawn from publicly available sources, acquired from third-parties, or synthetically generated or augmented. Publicly available data includes content from the public internet and publicly accessible repositories.","The training data curation process includes cleaning, processing, and modifying datasets. These processing steps, which vary by data type, may include deduplication and filtering to remove junk or other low-quality data, or to advance safety or other objectives.",[459,474,474,474,475,476,477,478,461,479],"# Call the server using curl (OpenAI-compatible API):","Inkling is a general-purpose multimodal model that accepts text, image and audio inputs and generates text outputs. It is intended for use in English and other languages, and across multiple coding languages. The model is designed to be used by developers building AI-powered applications, including agentic and tool-use systems, coding assistants, chatbots, and retrieval-augmented generation systems, and is suitable for general-purpose conversational use, instruction-following, and other natural language and multimodal tasks. It is released with open weights to support research, fine-tuning and integration into third-party products by downstream developers.","Languages: English, with general multilingual capabilities across other languages.","Try Inkling on the Tinker Playground or access via API using the Tinker Cookbook .","API access is also available through third party inference providers.","Conduct their own evaluation of Inkling's performance, safety, and fairness for their specific use case, language, and population prior to deployment, particularly for applications involving vulnerable groups.","thinkingmachines\u002FInkling · Hugging Face \nHugging Face Models Datasets Spaces Buckets new Docs Enterprise Pricing Website Tasks HuggingChat Collections Languages Organizations Community Blog Posts Daily Papers Hardware Learn Discord Forum GitHub Solutions Team & Enterprise Hugging Face PRO Enterprise Support Inference Providers Inference Endpoints Storage Buckets Log In Sign Up \\\", \\\"assistant\\\": \\\"\u003C|message_model|>\\\", \\\"system\\\": \\\"\u003C|message_system|>\\\", \\\"tool\\\": \\\"\u003C|message_tool|>\\\"} -%}\\n\\n{%- macro emit_thinking_effort() -%}\\n {%- set eff = reasoning_effort if reasoning_effort is defined and reasoning_effort is not none else 0.9 -%}\\n {%- if eff is string -%}\\n {%- set key = eff | trim -%}\\n {%- if key not in effort_map -%}\\n {{- raise_exception(\\\"Unknown reasoning_effort: \\\" ~ eff) -}}\\n {%- endif -%}\\n {%- set num = effort_map[key] -%}\\n {%- else -%}\\n {%- set num = eff | float -%}\\n {%- endif -%}\\n {%- if num \u003C 0.0 or num > 0.99 -%}\\n {{- raise_exception(\\\"reasoning_effort must be in [0.0, 0.99]\\\") -}}\\n {%- endif -%}\\n {{- \\\"\u003C|message_system|>\u003C|content_text|>Thinking effort level: \\\" -}}\\n {%- if num == 0.0 -%}0{%- else -%}{{ num }}{%- endif -%}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n{%- endmacro -%}\\n\\n{%- if tools -%}\\n {%- set tool_state = namespace(specs=[]) -%}\\n {%- for tool in tools -%}\\n {%- set fn = tool.function if tool.function is defined else tool -%}\\n {%- set spec = {\\n \\\"description\\\": (fn.description if fn.description is defined and fn.description else \\\"\\\"),\\n \\\"name\\\": fn.name,\\n \\\"parameters\\\": (fn.parameters if fn.parameters is defined and fn.parameters else {}),\\n \\\"type\\\": (tool.type if tool.type is defined and tool.type else \\\"function\\\"),\\n } -%}\\n {%- set tool_state.specs = tool_state.specs + [spec] -%}\\n {%- endfor -%}\\n {{- \\\"\u003C|message_system|>tool_declare\u003C|content_xml|>\\\" -}}\\n {{- tool_state.specs | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\")) -}}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n{%- endif -%}\\n\\n{%- set state = namespace(effort_emitted=false) -%}\\n{%- for message in messages -%}\\n {%- if message.role not in role_token -%}\\n {{- raise_exception(\\\"Unknown message role: \\\" ~ message.role) -}}\\n {%- endif -%}\\n {%- if not state.effort_emitted and message.role != \\\"system\\\" -%}\\n {{- emit_thinking_effort() -}}\\n {%- set state.effort_emitted = true -%}\\n {%- endif -%}\\n\\n {%- set rtok = role_token[message.role] -%}\\n\\n {%- if message.role == \\\"tool\\\" -%}\\n {%- set tool_name_state = namespace(name=\\\"\\\") -%}\\n {%- if message.name is defined and message.name -%}\\n {%- set tool_name_state.name = message.name -%}\\n {%- elif message.tool_call_id is defined and message.tool_call_id -%}\\n {%- for prev in messages -%}\\n {%- if prev.role == \\\"assistant\\\" and prev.tool_calls -%}\\n {%- for tc in prev.tool_calls -%}\\n {%- if tc.id is defined and tc.id == message.tool_call_id and tc.function.name is defined -%}\\n {%- set tool_name_state.name = tc.function.name -%}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n {{- rtok -}}\\n {%- if tool_name_state.name -%}{{- tool_name_state.name -}}{%- endif -%}\\n {{- \\\"\u003C|content_text|>\\\" -}}\\n {%- if message.content is string -%}{{- message.content -}}{%- endif -%}\\n {{- \\\"\u003C|end_message|>\\\" -}}\\n\\n {%- else -%}\\n {%- if message.role == \\\"assistant\\\" and message.reasoning_content is defined and message.reasoning_content -%}\\n {{- \\\"\u003C|message_model|>\u003C|content_thinking|>\\\" ~ message.reasoning_content ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- endif -%}\\n\\n {%- if message.content is string -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ message.content ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif message.content -%}\\n {%- for part in message.content -%}\\n {%- if part is string -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ part ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif part.type is not defined or part.type in (\\\"text\\\", \\\"input_text\\\") -%}\\n {%- set text_part = (part.text if part.text is defined and part.text is string else \\\"\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_text|>\\\" ~ text_part ~ \\\"\u003C|end_message|>\\\" -}}\\n {%- elif part.type in (\\\"image\\\", \\\"input_image\\\", \\\"image_url\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_image|>\u003C|unused_200054|>\u003C|end_message|>\\\" -}}\\n {%- elif part.type in (\\\"audio\\\", \\\"input_audio\\\", \\\"audio_url\\\") -%}\\n {{- rtok ~ \\\"\u003C|content_audio_input|>\u003C|unused_200053|>\u003C|audio_end|>\u003C|end_message|>\\\" -}}\\n {%- else -%}\\n {{- raise_exception(\\\"Unsupported content part type: \\\" ~ part.type) -}}\\n {%- endif -%}\\n {%- endfor -%}\\n {%- endif -%}\\n\\n {%- if message.role == \\\"assistant\\\" and message.tool_calls -%}\\n {%- for tc in message.tool_calls -%}\\n {%- set fn = tc.function -%}\\n {%- if fn.name is not defined or fn.name is not string -%}\\n {{- raise_exception(\\\"tool call function name must be a string\\\") -}}\\n {%- endif -%}\\n {%- set args = fn.arguments if fn.arguments is defined and fn.arguments else {} -%}\\n {%- if args is string -%}\\n {{- raise_exception(\\\"tool call arguments must be a parsed object, not a JSON string; canonicalize upstream\\\") -}}\\n {%- endif -%}\\n {%- if args is not mapping -%}\\n {{- raise_exception(\\\"tool call arguments must be an object\\\") -}}\\n {%- endif -%}\\n {{- \\\"\u003C|message_model|>\\\" ~ fn.name ~ \\\"\u003C|content_invoke_tool_json|>\\\" -}}\\n {{- '{\\\"name\\\":' ~ (fn.name | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\"))) ~ ',\\\"args\\\":' -}}\\n {{- (args | tojson(sort_keys=true, separators=(\\\",\\\", \\\":\\\"))) -}}\\n {{- \\\"}\u003C|end_message|>\\\" -}}\\n {%- endfor -%}\\n {%- endif -%}\\n\\n {%- if message.role == \\\"assistant\\\" -%}\\n {{- \\\"\u003C|content_model_end_sampling|>\\\" -}}\\n {%- endif -%}\\n {%- endif -%}\\n{%- endfor -%}\\n\\n{%- if not state.effort_emitted -%}\\n {{- emit_thinking_effort() -}}\\n{%- endif -%}\\n\\n{%- if add_generation_prompt -%}\\n {{- \\\"\u003C|message_model|>\\\" -}}\\n{%- endif -%}\\n\"},\"createdAt\":\"2026-07-14T13:23:14.000Z\",\"discussionsDisabled\":false,\"discussionsSorting\":\"recently-created\",\"downloads\":77799,\"downloadsAllTime\":77799,\"id\":\"thinkingmachines\u002FInkling\",\"isLikedByUser\":false,\"availableInferenceProviders\":[{\"provider\":\"baseten\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002Finkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":true,\"toolCalling\":true},\"isCheapestPricingOutput\":true,\"isFastestThroughput\":true,\"isModelAuthor\":false,\"tokensPerSecond\":158.72110743627016,\"pricingOutput\":4.05},{\"provider\":\"fireworks-ai\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"accounts\u002Ffireworks\u002Fmodels\u002Finkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":false,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":102.38685589674759},{\"provider\":\"together\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002FInkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":true,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":127.45434889475416,\"pricingOutput\":4.05},{\"provider\":\"deepinfra\",\"modelStatus\":\"live\",\"providerStatus\":\"live\",\"providerId\":\"thinkingmachines\u002FInkling\",\"task\":\"conversational\",\"adapterWeightsPath\":\"model-00001-of-00108.safetensors\",\"features\":{\"structuredOutput\":false,\"toolCalling\":true},\"isCheapestPricingOutput\":false,\"isFastestThroughput\":false,\"isModelAuthor\":false,\"tokensPerSecond\":99.40961150436104,\"pricingOutput\":4.05}],\"showHuggingChatEntry\":true,\"inference\":\"warm\",\"lastModified\":\"2026-07-23T17:27:17.000Z\",\"likes\":1703,\"pipeline_tag\":\"image-text-to-text\",\"library_name\":\"transformers\",\"librariesOther\":[],\"trackDownloads\":true,\"model-index\":null,\"evalResults\":[{\"dataset\":{\"id\":\"SWE-bench\u002FSWE-bench_Verified\",\"isBenchmark\":true,\"task_id\":\"swe_bench_%_resolved\"},\"value\":77.6,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fswe-bench_verified.yaml\",\"verified\":false,\"rank\":10,\"label\":\"Swe Bench Resolved\"},{\"dataset\":{\"id\":\"ScaleAI\u002FSWE-bench_Pro\",\"isBenchmark\":true,\"task_id\":\"SWE_Bench_Pro\"},\"value\":54.3,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fswe-bench_pro.yaml\",\"verified\":false,\"rank\":16,\"label\":\"SWE Bench Pro\"},{\"dataset\":{\"id\":\"Idavidrein\u002Fgpqa\",\"isBenchmark\":true,\"task_id\":\"diamond\"},\"value\":87.2,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fgpqa.yaml\",\"verified\":false,\"rank\":19,\"label\":\"Diamond\"},{\"dataset\":{\"id\":\"MathArena\u002Faime_2026\",\"isBenchmark\":true,\"task_id\":\"MathArena\u002Faime_2026\"},\"value\":97.1,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Faime_2026.yaml\",\"verified\":false,\"rank\":1,\"label\":\"MathArena Aime 2026\"},{\"dataset\":{\"id\":\"cais\u002Fhle\",\"isBenchmark\":false,\"task_id\":\"hle\"},\"value\":46,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fhle.yaml\",\"verified\":false,\"rank\":14,\"label\":\"Hle\"},{\"dataset\":{\"id\":\"MMMU\u002FMMMU_Pro\",\"isBenchmark\":true,\"task_id\":\"mmmu_pro_standard_10_options\"},\"value\":73.5,\"source\":{\"url\":\"https:\u002F\u002Fhuggingface.co\u002Fthinkingmachines\u002FInkling\",\"name\":\"Model Card\",\"isExternal\":false},\"filename\":\".eval_results\u002Fmmmu_pro.yaml\",\"verified\":false,\"rank\":3,\"label\":\"Mmmu Pro Standard 10 Options\"}],\"private\":false,\"repoType\":\"model\",\"gated\":false,\"tags\":[\"transformers\",\"safetensors\",\"inkling_mm_model\",\"image-text-to-text\",\"conversational\",\"audio-text-to-text\",\"moe\",\"license:apache-2.0\",\"eval-results\",\"endpoints_compatible\",\"region:us\"],\"tag_objs\":[{\"id\":\"image-text-to-text\",\"label\":\"Image-Text-to-Text\",\"type\":\"pipeline_tag\",\"subType\":\"multimodal\"},{\"id\":\"transformers\",\"label\":\"Transformers\",\"type\":\"library\"},{\"id\":\"safetensors\",\"label\":\"Safetensors\",\"type\":\"library\"},{\"id\":\"inkling_mm_model\",\"label\":\"inkling_mm_model\",\"type\":\"other\",\"clickable\":true},{\"id\":\"conversational\",\"label\":\"conversational\",\"type\":\"other\",\"clickable\":true},{\"id\":\"audio-text-to-text\",\"label\":\"audio-text-to-text\",\"type\":\"other\",\"clickable\":true},{\"id\":\"moe\",\"label\":\"Mixture of Experts\",\"type\":\"other\",\"clickable\":true},{\"id\":\"eval-results\",\"label\":\"Eval Results\",\"type\":\"other\",\"clickable\":true},{\"id\":\"endpoints_compatible\",\"label\":\"Inference Endpoints\",\"type\":\"other\",\"clickable\":true},{\"id\":\"license:apache-2.0\",\"label\":\"apache-2.0\",\"type\":\"license\"},{\"type\":\"region\",\"label\":\"🇺🇸 Region: US\",\"id\":\"region:us\"}],\"transformersInfo\":{\"auto_model\":\"AutoModelForMultimodalLM\",\"pipeline_tag\":\"image-text-to-text\",\"processor\":\"AutoProcessor\"},\"widgetData\":[{\"text\":\"Hi, what can you help me with?\"},{\"text\":\"What is 84 * 3 \u002F 2?\"},{\"text\":\"Tell me an interesting fact about the universe!\"},{\"text\":\"Explain quantum computing in simple terms.\"}],\"safetensors\":{\"parameters\":{\"BF16\":952377607178,\"F32\":16448},\"total\":952377623626,\"sharded\":true,\"totalFileSize\":1904755592540},\"hasBlockedOids\":false,\"region\":\"us\",\"isQuantized\":false},\"discussionsStats\":{\"closed\":17,\"open\":0,\"total\":17},\"query\":{},\"inferenceContextData\":{\"billableEntities\":[],\"entityName2Providers\":{},\"defaultProviders\":[{\"isOriginalProvider\":false,\"name\":\"together\",\"enabled\":true,\"position\":5,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"fireworks-ai\",\"enabled\":true,\"position\":6,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"baseten\",\"enabled\":true,\"position\":13,\"isReleased\":true,\"accuratePricing\":true},{\"isOriginalProvider\":false,\"name\":\"deepinfra\",\"enabled\":true,\"position\":15,\"isReleased\":true,\"accuratePricing\":true}]},\"hasQuantizations\":true,\"copyToBucketNamespaces\":[]}\"> thinkingmachines \u002F Inkling like 1.7k Follow Thinking Machines Lab 1.65k \nImage-Text-to-Text Transformers Safetensors inkling_mm_model conversational audio-text-to-text Mixture of Experts Eval Results License: apache-2.0 Model card Files Files and versions xet Community 17 Deploy Copy to bucket new Use this model Instructions to use thinkingmachines\u002FInkling with libraries, inference providers, notebooks, and local apps. Follow these links to get started.\nLibraries Transformers How to use thinkingmachines\u002FInkling with Transformers:\n# Use a pipeline as a high-level helper\nfrom transformers import pipeline\npipe = pipeline(\"image-text-to-text\", model=\"thinkingmachines\u002FInkling\")\nmessages = [\n{\n\"role\": \"user\",\n\"content\": [\n{\"type\": \"image\", \"url\": \"https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fhuggingface\u002Fdocumentation-images\u002Fresolve\u002Fmain\u002Fp-blog\u002Fcandy.JPG\"},\n{\"type\": \"text\", \"text\": \"What animal is on the candy?\"}\n]\n},\n]\npipe(text=messages) # Load model directly\nfrom transformers import AutoProcessor, AutoModelForMultimodalLM\nprocessor = AutoProcessor.from_pretrained(\"thinkingmachines\u002FInkling\")\nmodel = AutoModelForMultimodalLM.from_pretrained(\"thinkingmachines\u002FInkling\", device_map=\"auto\")\nmessages = [\n{\n\"role\": \"user\",\n\"content\": [\n{\"type\": \"image\", \"url\": \"https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fhuggingface\u002Fdocumentation-images\u002Fresolve\u002Fmain\u002Fp-blog\u002Fcandy.JPG\"},\n{\"type\": \"text\", \"text\": \"What animal is on the candy?\"}\n]\n},\n]\ninputs = processor.apply_chat_template(\nmessages,\nadd_generation_prompt=True,\ntokenize=True,\nreturn_dict=True,\nreturn_tensors=\"pt\",\n).to(model.device)\noutputs = model.generate(**inputs, max_new_tokens=40)\nprint(processor.decode(outputs[0][inputs[\"input_ids\"].shape[-1]:])) Inference Inference Providers HuggingChat Notebooks Google Colab Kaggle Local Apps Settings vLLM How to use thinkingmachines\u002FInkling with vLLM:\nInstall from pip and serve model\n# Install vLLM from pip:\npip install vllm\n# Start the vLLM server:\nvllm serve \"thinkingmachines\u002FInkling\"\n# Call the server using curl (OpenAI-compatible API):\ncurl -X POST \"http:\u002F\u002Flocalhost:8000\u002Fv1\u002Fchat\u002Fcompletions\" \\\n-H \"Content-Type: application\u002Fjson\" \\\n--data '{\n\"model\": \"thinkingmachines\u002FInkling\",\n\"messages\": [\n{\n\"role\": \"user\",\n\"content\": [\n{\n\"type\": \"text\",\n\"text\": \"Describe this image in one sentence.\"\n},\n{\n\"type\": \"image_url\",\n\"image_url\": {\n\"url\": \"https:\u002F\u002Fcdn.britannica.com\u002F61\u002F93061-050-99147DCE\u002FStatue-of-Liberty-Island-New-York-Bay.jpg\"\n}\n}\n]\n}\n]\n}' Use Docker\ndocker model run hf.co\u002Fthinkingmachines\u002FInkling SGLang How to use thinkingmachines\u002FInkling with SGLang:\nInstall from pip and serve model\n# Install SGLang from pip:\npip install sglang\n# Start the SGLang server:\npython3 -m sglang.launch_server \\\n--model-path \"thinkingmachines\u002FInkling\" \\\n--host 0.0.0.0 \\\n--port 30000\n# Call the server using curl (OpenAI-compatible API):\ncurl -X POST \"http:\u002F\u002Flocalhost:30000\u002Fv1\u002Fchat\u002Fcompletions\" \\\n-H \"Content-Type: application\u002Fjson\" \\\n--data '{\n\"model\": \"thinkingmachines\u002FInkling\",\n\"messages\": [\n{\n\"role\": \"user\",\n\"content\": [\n{\n\"type\": \"text\",\n\"text\": \"Describe this image in one sentence.\"\n},\n{\n\"type\": \"image_url\",\n\"image_url\": {\n\"url\": \"https:\u002F\u002Fcdn.britannica.com\u002F61\u002F93061-050-99147DCE\u002FStatue-of-Liberty-Island-New-York-Bay.jpg\"\n}\n}\n]\n}\n]\n}' Use Docker images\ndocker run --gpus all \\\n--shm-size 32g \\\n-p 30000:30000 \\\n-v ~\u002F.cache\u002Fhuggingface:\u002Froot\u002F.cache\u002Fhuggingface \\\n--env \"HF_TOKEN=\u003Csecret>\" \\\n--ipc=host \\\nlmsysorg\u002Fsglang:latest \\\npython3 -m sglang.launch_server \\\n--model-path \"thinkingmachines\u002FInkling\" \\\n--host 0.0.0.0 \\\n--port 30000\n# Call the server using curl (OpenAI-compatible API):\ncurl -X POST \"http:\u002F\u002Flocalhost:30000\u002Fv1\u002Fchat\u002Fcompletions\" \\\n-H \"Content-Type: application\u002Fjson\" \\\n--data '{\n\"model\": \"thinkingmachines\u002FInkling\",\n\"messages\": [\n{\n\"role\": \"user\",\n\"content\": [\n{\n\"type\": \"text\",\n\"text\": \"Describe this image in one sentence.\"\n},\n{\n\"type\": \"image_url\",\n\"image_url\": {\n\"url\": \"https:\u002F\u002Fcdn.britannica.com\u002F61\u002F93061-050-99147DCE\u002FStatue-of-Liberty-Island-New-York-Bay.jpg\"\n}\n}\n]\n}\n]\n}' Docker Model Runner How to use thinkingmachines\u002FInkling with Docker Model Runner:\ndocker model run hf.co\u002Fthinkingmachines\u002FInkling Browse\nQuantizations to use this model in llama.cpp , Ollama , LM Studio , or any compatible app. Inkling 1. General Information 2. Getting Started 3. Model Properties Model type Architecture type Parameters Numerics support Input modalities Output modalities 4. Training 5. Evaluations 6. Safety 7. Bias, risks and limitations \nInkling\nBF16 |\nNVFP4 |\nPlayground |\nTinker Cookbook |\nAcceptable Use \n1. General Information\nInkling is a general-purpose multimodal model that accepts text, image and audio inputs and generates text outputs. It is intended for use in English and other languages, and across multiple coding languages. The model is designed to be used by developers building AI-powered applications, including agentic and tool-use systems, coding assistants, chatbots, and retrieval-augmented generation systems, and is suitable for general-purpose conversational use, instruction-following, and other natural language and multimodal tasks. It is released with open weights to support research, fine-tuning and integration into third-party products by downstream developers.\nLanguages: English, with general multilingual capabilities across other languages.\n2. Getting Started\nTry Inkling on the Tinker Playground or access via API using the Tinker Cookbook .\nInkling supports local deployment using the following open-source libraries:\nSGLang ( recipe ) \nvLLM ( recipe ) \nTokenSpeed ( recipe ) \nUnsloth ( recipe ) \nHuggingface ( recipe ) \nAPI access is also available through third party inference providers.\n3. Model Properties\nModel type\nMultimodal autoregressive transformer\nArchitecture type\nA 66-layer decoder-only transformer with a sparse Mixture-of-Experts (MoE) feed-forward backbone: each token is routed to 6 of 256 experts, plus 2 shared experts active on every token. Attention is a hybrid of local and global layers. The model is natively multimodal — images and video are encoded via a hierarchical patch encoder, and audio via discrete token encoding — with all modalities projected into a shared hidden",[482,499,516,561,596,611,643,662],{"url":483,"kind":484,"anchor":485,"title":486,"description":437,"headings":487,"feature_evidence":489,"use_case_evidence":496,"fetched_at":497,"status":425,"content_hash":498},"https:\u002F\u002Fhuggingface.co\u002Fdocs","feature","Docs","Hugging Face - Documentation",[488],"Documentation",[490,491,492,493,494,495],"Hugging Face Models Datasets Spaces Buckets new Docs Enterprise Pricing Website Tasks HuggingChat Collections Languages Organizations Community Blog Posts Daily Papers Hardware Learn Discord Forum GitHub Solutions Team & Enterprise Hugging Face PRO Enterprise Support Inference Providers Inference Endpoints Storage Buckets Log In Sign Up Documentation","Guides, references, and API docs for the Hugging Face ecosystem.","API for metadata, stats, and content of HF Hub datasets","Serve language models with TGI optimized toolkit","Parameter-efficient finetuning for large language models","AutoTrain API and UI for seamless model training",[490],"2026-08-09T01:07:44.900Z","990104750e302df42cfbe57c0e9bda24a99fe40a002a982fabe2c01f891a38ce",{"url":500,"kind":484,"anchor":352,"title":501,"description":437,"headings":502,"feature_evidence":510,"use_case_evidence":513,"fetched_at":514,"status":425,"content_hash":515},"https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Fsafetensors","Safetensors · Hugging Face",[503,503,504,505,506,507,508,509],"Safetensors","Installation","Usage","Load tensors","Save tensors","Format","Featured Projects",[511,512],"Hugging Face Models Datasets Spaces Buckets new Docs Enterprise Pricing Website Tasks HuggingChat Collections Languages Organizations Community Blog Posts Daily Papers Hardware Learn Discord Forum GitHub Solutions Team & Enterprise Hugging Face PRO Enterprise Support Inference Providers Inference Endpoints Storage Buckets Log In Sign Up Safetensors documentation","🏡 View all docs AWS Trainium & Inferentia Accelerate Argilla AutoTrain Bitsandbytes CLI Chat UI Dataset viewer Datasets Deploying on AWS Diffusers Distilabel Evaluate Google Cloud Google TPUs Gradio Hub Hub Python Library Huggingface.js Inference Endpoints (dedicated) Inference Providers Kernels LeRobot Leaderboards Lighteval Microsoft Azure OpenEnv Optimum PEFT Reachy Mini Safetensors Sentence Transformers TRL Tasks Text Embeddings Inference Text Generation Inference Tokenizers Trackio Transformers Transformers.js Xet smolagents timm Search documentation main v0.5.0-rc.0 v0.3.2 v0.2.9 EN Getting started 🤗 Safetensors Speed Comparison Tensor Sharing in Pytorch Metadata Parsing Convert weights to safetensors API Torch API Tensorflow API PaddlePaddle API Flax API Numpy API You are viewing main version, which requires installation from source . If you'd like",[511],"2026-08-09T01:07:45.451Z","c698daac80d92308eba5b562cc1bf12e1cba86f6ef0b3e5bad9251fbae968243",{"url":517,"kind":484,"anchor":518,"title":519,"description":437,"headings":520,"feature_evidence":538,"use_case_evidence":557,"fetched_at":559,"status":425,"content_hash":560},"https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Finference-providers","NEW","Inference Providers · Hugging Face",[521,521,522,523,524,525,526,527,528,529,530,531,532,533,530,531,534,535,536,537],"Inference Providers","Quick Setup for Agents","Partners","Why Choose Inference Providers?","Key Features","Getting Started","Inference Playground","Authentication","Quick Start - LLM","Python","JavaScript","HTTP \u002F cURL","Quick Start - Text-to-Image Generation","Provider Selection","API as a Proxy Service","Client-Side Provider Selection (Inference Clients)","Alternative: OpenAI-Compatible Chat Completions Endpoint (Chat Only)",[539,540,541,542,543,544,545,546,547,548,549,550,551,535,552,553,554,555,556],"Hugging Face Models Datasets Spaces Buckets new Docs Enterprise Pricing Website Tasks HuggingChat Collections Languages Organizations Community Blog Posts Daily Papers Hardware Learn Discord Forum GitHub Solutions Team & Enterprise Hugging Face PRO Enterprise Support Inference Providers Inference Endpoints Storage Buckets Log In Sign Up Inference Providers documentation","🏡 View all docs AWS Trainium & Inferentia Accelerate Argilla AutoTrain Bitsandbytes CLI Chat UI Dataset viewer Datasets Deploying on AWS Diffusers Distilabel Evaluate Google Cloud Google TPUs Gradio Hub Hub Python Library Huggingface.js Inference Endpoints (dedicated) Inference Providers Kernels LeRobot Leaderboards Lighteval Microsoft Azure OpenEnv Optimum PEFT Reachy Mini Safetensors Sentence Transformers TRL Tasks Text Embeddings Inference Text Generation Inference Tokenizers Trackio Transformers Transformers.js Xet smolagents timm Search documentation main EN Get Started Inference Providers Pricing and Billing Hub integration Security Guides Your First API Call Building Your First AI App Structured Outputs with LLMs Function Calling Responses API (beta) How to use OpenAI gpt-oss Build an Image Editor Automating Code Review with GitHub Actions Agentic Coding Environments with OpenEnv Evaluating Models with Inspect Integrations Overview Pi OpenCode Codex Claude Code Hermes Agent NeMo Data Designer MacWhisper Vision Agents VS Code with GitHub Copilot Add Your Integration Inference Tasks Chat Completion Feature Extraction Text to Image Text to Video Other Tasks Providers Baseten Cerebras Cohere DeepInfra Fal AI Featherless AI Fireworks Groq HF Inference Novita Nscale OVHcloud AI Endpoints Public AI Replicate Scaleway Together WaveSpeedAI Z.ai Hub API Register as an Inference Provider Join the Hugging Face community and get access to the augmented documentation experience","Our platform integrates with leading AI infrastructure providers, giving you access to their specialized capabilities through a single, consistent API. Here’s what each partner supports:","When you build AI applications, it’s tough to manage multiple provider APIs, comparing model performance, and dealing with varying reliability. Inference Providers solves these challenges by offering:","Instant Access to Cutting-Edge Models : Go beyond mainstream providers to access thousands of specialized models across multiple AI tasks. Whether you need the latest language models, state-of-the-art image generators, or domain-specific embeddings, you’ll find them here.","Text Generation : Use Large language models with tool-calling capabilities for chatbots, content generation, and code assistance Image and Video Generation : Create custom images and videos, including support for LoRAs and style customization Search & Retrieval : State-of-the-art embeddings for semantic search, RAG systems, and recommendation engines Traditional ML Tasks : Ready-to-use models for classification, NER, summarization, and speech recognition ⚡ Get Started for Free : Inference Providers includes a generous free tier, with additional credits for PRO users and Team & Enterprise organizations .","🎯 All-in-One API : A single API for text generation, image generation, document embeddings, NER, summarization, image classification, and more. 🔀 Multi-Provider Support : Easily run models from top-tier providers like fal, Replicate, Together AI, and others. 🚀 Scalable & Reliable : Built for high availability and low-latency performance in production environments. 🔧 Developer-Friendly : Simple requests, fast responses, and a consistent developer experience across Python and JavaScript clients. 👷 Easy to integrate : Drop-in replacement for the OpenAI chat completions API. 💰 Cost-Effective : No extra markup on provider rates. Getting Started","Inference Providers works with your existing development workflow. Whether you prefer Python, JavaScript, or direct HTTP calls, we provide native SDKs and OpenAI-compatible APIs to get you up and running quickly.","Prefer the terminal? Run hf models ls --warm with the hf CLI to list every model served by at least one provider, and add --json for scripts and agents. See the Hub API page for the full set of filters.","Let’s start with the most common use case: conversational AI using large language models. This section demonstrates how to perform chat completions using DeepSeek V3, showcasing the different ways you can integrate Inference Providers into your applications.","For testing, debugging, or integrating with any HTTP client, here’s the raw REST API format.","client = InferenceClient(api_key=os.environ[ \"HF_TOKEN\" ])","The Inference Providers API acts as a unified proxy layer that sits between your application and multiple AI providers. Understanding how provider selection works is crucial for optimizing performance, cost, and reliability in your applications.","Unified Authentication & Billing : Use a single Hugging Face token for all providers Automatic Failover : When using automatic provider selection ( provider=\"auto\" ), requests are automatically routed to alternative providers if the primary provider is flagged as unavailable by our validation system Consistent Interface through client libraries : When using our client libraries, the same request format works across different providers Because the API acts as a proxy, the exact HTTP request may vary between providers as each provider has their own API requirements and response formats. When using our official client libraries (JavaScript or Python), these provider-specific differences are handled automatically whether you use provider=\"auto\" or specify a particular provider.","When using the Hugging Face inference clients (JavaScript or Python), you can explicitly specify a provider or let the system choose automatically. The client then formats the HTTP request to match the selected provider’s API requirements.","If you prefer to work with familiar OpenAI APIs or want to migrate existing chat completion code with minimal changes, we offer a drop-in compatible endpoint that handles all provider selection automatically on the server side.","apiKey : process. env . HF_TOKEN ,","Announcement Blog Post : Learn more about the launch of Inference Providers Pricing and Billing : Understand costs and billing of Inference Providers Hub Integration : Learn how Inference Providers are integrated with the Hugging Face Hub Register as a Provider : Requirements to join our partner network as a provider Hub API : Advanced API features and configuration API Reference : Complete parameter documentation for all supported tasks Update on GitHub",[539,546,558,548],"Before diving into integration, explore models interactively with our Inference Playground . Test different chat completion models with your prompts and compare responses to find the perfect fit for your use case.","2026-08-09T01:07:45.885Z","b28dbeccbfc963ee5708fd8619a1a264fbe4e0b52183939a94677f56a883261c",{"url":562,"kind":484,"anchor":352,"title":563,"description":437,"headings":564,"feature_evidence":585,"use_case_evidence":593,"fetched_at":594,"status":425,"content_hash":595},"https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Fhub\u002Fmodel-cards#specifying-a-base-model","Model Cards · Hugging Face",[565,566,567,568,569,570,571,572,573,574,575,576,577,578,579,580,581,582,583,584],"Hub","Model Cards","What are Model Cards?","Model card metadata","Adding metadata to your model card","Using the metadata UI","Editing the YAML section of the README.md file","Specifying a library","Specifying a base model","Specifying a new version","Specifying a dataset","Specifying a bucket","Specifying a task ( pipeline_tag )","Specifying a license","Evaluation Results","CO2 Emissions","Linking a Paper","Model Card text","FAQ","How are model tags determined?",[586,587,588,589,590,591,592],"Hugging Face Models Datasets Spaces Buckets new Docs Enterprise Pricing Website Tasks HuggingChat Collections Languages Organizations Community Blog Posts Daily Papers Hardware Learn Discord Forum GitHub Solutions Team & Enterprise Hugging Face PRO Enterprise Support Inference Providers Inference Endpoints Storage Buckets Log In Sign Up Hub documentation","🏡 View all docs AWS Trainium & Inferentia Accelerate Argilla AutoTrain Bitsandbytes CLI Chat UI Dataset viewer Datasets Deploying on AWS Diffusers Distilabel Evaluate Google Cloud Google TPUs Gradio Hub Hub Python Library Huggingface.js Inference Endpoints (dedicated) Inference Providers Kernels LeRobot Leaderboards Lighteval Microsoft Azure OpenEnv Optimum PEFT Reachy Mini Safetensors Sentence Transformers TRL Tasks Text Embeddings Inference Text Generation Inference Tokenizers Trackio Transformers Transformers.js Xet smolagents timm Search documentation EN API docs 🤗 Hugging Face Hub Team & Enterprise Plans Single Sign-On (SSO) Audit Logs Storage Regions Data Studio for Private datasets Resource Groups (Access Control) Advanced Compute Options Advanced Security Tokens Management Service Accounts Publisher Analytics Gating Group Collections Network Security Download Analytics Rate Limits Blog Articles PRO Plan Repositories Getting Started with Repositories Repository Settings Storage Limits Storage Backend (Xet) Local Cache Pull Requests & Discussions Notifications Collections Webhooks GitHub Actions Notebooks Next Steps Licenses Models The Model Hub Model Cards Carbon Emissions Card Components Eval Results Leaderboard Data Gated Models Uploading Models Downloading Models Integrated Libraries Model Widgets Model Inference Models Download Stats Model Release Checklist Hardware Local Apps Frequently Asked Questions Advanced Topics Datasets Datasets Overview Dataset Cards Gated Datasets Uploading Datasets Uploading Datasets (for LLMs) Downloading Datasets Streaming Datasets Integrated Libraries Data Studio Agent Traces Datasets Download Stats Spaces Spaces Overview Spaces GPU Upgrades Spaces ZeroGPU Spaces Dev Mode Spaces Disk Usage & Storage Spaces Custom Domain Spaces as MCP servers Spaces as Agent Tools Spaces as API Endpoints Gradio Spaces Streamlit Spaces Static HTML Spaces Docker Spaces Embed your Space Run Spaces with Docker Spaces Configuration Reference Sign-In with HF button Featured Spaces Spaces Changelog Advanced Topics Storage Buckets new Access Patterns S3 Compatibility Bucket Integrations Bucket Security Jobs Jobs Overview Quickstart Pricing and Billing Manage Jobs Configuration Popular Images Serve Models Examples & Tutorials Process Large Datasets Schedule Jobs Webhook Automation Reference Agents Agents Overview Hugging Face CLI for AI Agents Hugging Face MCP Server Hugging Face Agent Skills Building agents with the HF SDK Local Agents with llama.cpp Agent Libraries Session Traces Format Other Organizations Billing Security Moderation Paper Pages Academia Hub Blog Articles Search Digital Object Identifier (DOI) Hub API Endpoints OAuth \u002F Sign in with HF Join the Hugging Face community and get access to the augmented documentation experience","the model its intended uses & potential limitations, including biases and ethical considerations as detailed in Mitchell, 2018 the training params and experimental info (you can embed or link to an experiment tracking platform for reference) which datasets were used to train your model the model’s evaluation results The model card template is available here .","Allowing users to filter models at https:\u002F\u002Fhuggingface.co\u002Fmodels . Displaying the model’s license. Adding datasets to the metadata will add a message reading Datasets used to train: to your model page and link the relevant datasets, if they’re available on the Hub. Dataset and language identifiers are those listed on the Datasets and Languages pages.","language:","- \"List of ISO 639-1 code for your language\"","You can specify the pipeline_tag in the model card metadata. The pipeline_tag indicates the type of task the model is intended for. This tag will be displayed on the model page and users can filter models on the Hub by task. This tag is also used to determine which widget to use for the model and which APIs to use under the hood.",[586],"2026-08-09T01:07:46.209Z","4a2649199209784210d6ea49562458af1f67c515f1dc756887ad90e4fc72a4c6",{"url":597,"kind":484,"anchor":352,"title":598,"description":437,"headings":599,"feature_evidence":606,"use_case_evidence":608,"fetched_at":609,"status":425,"content_hash":610},"https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Fhub\u002Feval-results","Evaluation Results · Hugging Face",[565,579,600,601,602,603,604,605],"Benchmark Datasets","Model Evaluation Results","Adding Evaluation Results","Community Contributions","Registering a Benchmark","Eval.yaml specification",[586,607],"🏡 View all docs AWS Trainium & Inferentia Accelerate Argilla AutoTrain Bitsandbytes CLI Chat UI Dataset viewer Datasets Deploying on AWS Diffusers Distilabel Evaluate Google Cloud Google TPUs Gradio Hub Hub Python Library Huggingface.js Inference Endpoints (dedicated) Inference Providers Kernels LeRobot Leaderboards Lighteval Microsoft Azure OpenEnv Optimum PEFT Reachy Mini Safetensors Sentence Transformers TRL Tasks Text Embeddings Inference Text Generation Inference Tokenizers Trackio Transformers Transformers.js Xet smolagents timm Search documentation EN API docs 🤗 Hugging Face Hub Team & Enterprise Plans Single Sign-On (SSO) Audit Logs Storage Regions Data Studio for Private datasets Resource Groups (Access Control) Advanced Compute Options Advanced Security Tokens Management Service Accounts Publisher Analytics Gating Group Collections Network Security Download Analytics Rate Limits Blog Articles PRO Plan Repositories Getting Started with Repositories Repository Settings Storage Limits Storage Backend (Xet) Local Cache Pull Requests & Discussions Notifications Collections Webhooks GitHub Actions Notebooks Next Steps Licenses Models The Model Hub Model Cards Eval Results Leaderboard Data Gated Models Uploading Models Downloading Models Integrated Libraries Model Widgets Model Inference Models Download Stats Model Release Checklist Hardware Local Apps Frequently Asked Questions Advanced Topics Datasets Datasets Overview Dataset Cards Gated Datasets Uploading Datasets Uploading Datasets (for LLMs) Downloading Datasets Streaming Datasets Integrated Libraries Data Studio Agent Traces Datasets Download Stats Spaces Spaces Overview Spaces GPU Upgrades Spaces ZeroGPU Spaces Dev Mode Spaces Disk Usage & Storage Spaces Custom Domain Spaces as MCP servers Spaces as Agent Tools Spaces as API Endpoints Gradio Spaces Streamlit Spaces Static HTML Spaces Docker Spaces Embed your Space Run Spaces with Docker Spaces Configuration Reference Sign-In with HF button Featured Spaces Spaces Changelog Advanced Topics Storage Buckets new Access Patterns S3 Compatibility Bucket Integrations Bucket Security Jobs Jobs Overview Quickstart Pricing and Billing Manage Jobs Configuration Popular Images Serve Models Examples & Tutorials Process Large Datasets Schedule Jobs Webhook Automation Reference Agents Agents Overview Hugging Face CLI for AI Agents Hugging Face MCP Server Hugging Face Agent Skills Building agents with the HF SDK Local Agents with llama.cpp Agent Libraries Session Traces Format Other Organizations Billing Security Moderation Paper Pages Academia Hub Blog Articles Search Digital Object Identifier (DOI) Hub API Endpoints OAuth \u002F Sign in with HF Join the Hugging Face community and get access to the augmented documentation experience",[586],"2026-08-09T01:07:46.640Z","56903532887cc529d57f773b5b03a03571e7b8d2432993ef10249939d1502d20",{"url":612,"kind":613,"anchor":614,"title":615,"description":616,"headings":617,"feature_evidence":637,"use_case_evidence":640,"fetched_at":641,"status":425,"content_hash":642},"https:\u002F\u002Fhuggingface.co\u002Fpapers","research","Daily Papers","Daily Papers - Hugging Face","Your daily dose of AI research from AK",[614,618,619,620,621,622,623,624,625,626,627,628,629,630,631,632,633,634,635,636],"by AK and the research community","AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning","OSReward: Instituting Standardized Evaluation for Cross-Platform Computer-Use Reward Models","Interpretable MEG Decoding of Perceived Speech: Cortical Sources and the Stimulus Features That Drive Retrieval","WorldClaw: Agentic 3D Open-World Generation at Scale","GST-Bench: Can VLMs Develop Global Spatial Awareness from Video?","EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning","ChronoVision: Temporal Reasoning via Latent State Reconstruction","Learning from Failures: Retrieval-Centric CoT via Hard Negatives for Unified Multimodal Retrieval","HarnessOpt-Bench: Evaluating LLMs at Harness Optimization","From Economic Agents to Agentic Economies: A Systems Blueprint for Economic World Models","DataSpace: Benchmarking Data Agents for Verifiable Analytics over Heterogeneous Workspaces","On-Policy Delta Distillation for Multilingual Math Reasoning","Teaching Nemotron Greek: Mining a Corpus, Adapting Retrieval, and Grounding Generation for Modern Greek across Specialist Domains","World-to-Wrist: Task-Conditioned Future Wrist Modeling for Fine-Grained Robot Manipulation","DyPES-VLA: Learning Shared Dynamics Priors and Embodiment-Specific Control for Cross-Embodiment Manipulation","Activity Frames: Deterministic Screen-Activity Compilation for Agent Memory and Replay","SmartMage: Dynamic Modality Orchestration for 3D Scene Understanding","EffectLearner: World-Aware Object-Effect Reasoning for Real-World Video Object Removal",[638,639],"Hugging Face Models Datasets Spaces Buckets new Docs Enterprise Pricing Website Tasks HuggingChat Collections Languages Organizations Community Blog Posts Daily Papers Hardware Learn Discord Forum GitHub Solutions Team & Enterprise Hugging Face PRO Enterprise Support Inference Providers Inference Endpoints Storage Buckets Log In Sign Up new Get trending papers in your email inbox once a day!","Kandinsky Lab 53 3 Submitted by Uri-ka 14 MameLoshnLM: Yiddish Language Model and Evaluation Benchmark",[638],"2026-08-09T01:07:47.106Z","81ef1f8e47d94fcea90328e6f9de0dfd619225623f7b00abb0d70324368cf872",{"url":644,"kind":645,"title":644,"description":352,"feature_evidence":646,"use_case_evidence":659,"fetched_at":660,"status":425,"content_hash":661},"https:\u002F\u002Fhuggingface.co\u002Fsitemap-blog.xml","blog",[647,648,649,650,651,652,653,654,655,656,657,658],"https:\u002F\u002Fhuggingface.co\u002Fblog\u002Frapidfireai","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fanylanguagemodel","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fintel-protein-language-model-protst","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Ftgi-messages-api","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Frun-musicgen-as-an-api","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Funity-api","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fchinese-language-blog","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fintel-sapphire-rapids-inference","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fvision_language_pretraining","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fintel-sapphire-rapids","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Flarge-language-models","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Ffew-shot-learning-gpt-neo-and-inference-api",[],"2026-08-09T01:07:47.450Z","6f233b122ad09e75c57f6654c77242f0f50f9b12865961ac10367d14d836bb82",{"url":663,"kind":613,"title":663,"description":352,"feature_evidence":664,"use_case_evidence":665,"fetched_at":666,"status":425,"content_hash":667},"https:\u002F\u002Fhuggingface.co\u002Fsitemap-papers.xml",[],[],"2026-08-09T01:07:48.311Z","ad2833e02848abff8ab6f61e4988203adb61395a39ac87b4aad5879cda9bddf4",[669,670],{"url":612,"title":614,"kind":613},{"url":663,"title":663,"kind":613},[672],{"title":644,"url":644,"fetched_at":660},[],[459,474,474,474,475,476,477,478,461,479,490,491,492,493,494,495,511,512,539,540,541,542,543,544,545,546,547,548,549,550,551,535,552,553,554,555,556,586,587,588,589,590,591,592,586,607],"yes",[677],{"fetched_at":678,"title":679,"url":372},"2026-07-23T17:27:17.000Z","Introducing Inkling: A New Era of Open-Weights AI","unknown","Tak, model był pretrenowany na 45 bilionach tokenów tekstu, obrazów, audio i wideo z danych publicznie dostępnych, od stron trzecich oraz syntetycznie generowanych\u002Faugmentowanych.","model",[684,685,686,687,688,689,690,691,692,693,694,695,696,697,698,699,700,701],"Natywna multimodalność: tekst, obraz i audio jako wejście, tekst jako wyjście.","Okno kontekstu do 1M tokenów.","Kontrolowany 'thinking effort' pozwalający balansować jakość i koszt tokenów.","Architektura Mixture-of-Experts z 975B parametrów całkowitych i 41B aktywnych.","Wsparcie dla fine-tuningu przez Tinker.","Dostęp przez API partnerów inference oraz przez Hugging Face weights.","Wsparcie lokalnego wdrożenia w SGLang, vLLM, TokenSpeed, Unsloth i Hugging Face Transformers.","Narzędzia do pracy z audio i obrazami, w tym cookbook i playground.","multimodal input (text, images, audio)","output in text and code","context window of up to 1 million tokens","adjustable reasoning effort (fine-tuning of thinking depth)","Mixture of Experts (MoE) architecture","support for tool calling","structured output","available in multiple quantization formats (GGUF, BF16, NVFP4)","fine-tuning available via Tinker","open weights under Apache 2.0 license","Yes, the model weights are freely downloadable on Hugging Face under the Apache 2.0 license.",[704,705,706,707,708,709,710,711,712,713,714,715,367,716,717,718],"Tinker Playground","Tinker Cookbook","TogetherAI","Fireworks","Modal","Databricks","Baseten","SGLang","vLLM","TokenSpeed","Unsloth","Hugging Face Transformers","Tinker","Together AI","Fireworks AI",[720,721],"English","other languages","Apache 2.0",[724,725,726,727,728,729,730,731],"Wydajność może być nierówna między językami, dialektami i domenami słabiej reprezentowanymi w treningu.","Ograniczona wiedza do momentu cutoff treningu.","Może generować nieprawdziwe lub niepodparte odpowiedzi.","Producent zaleca dodatkowe zabezpieczenia i nadzór w zastosowaniach wysokiego ryzyka.","Outputs are limited to text and code; no image generation","Performance is lower than top-tier proprietary models","Full model requires large GPU clusters","Only available in specific quantized formats for local inference",[305,733],"Inkling-Small","Nie wykazano",[736],{"plan_name":737,"access_model":738,"price":739,"currency":740,"billing_period":741,"included_usage":680,"trial_days":680,"card_required":742,"source_evidence":433},"Inference via API partners","open_weights","4.05","USD","per_token","no","Brak zweryfikowanych oficjalnych cen startowych i planów cenowych w zebranym materiale. Producent podaje jedynie, że Inkling jest dostępny na Tinker z 50% zniżką przez ograniczony czas, a pełne ceny są w dokumentacji.","Nie znaleziono potwierdzenia, aby Inkling był trenowany na danych użytkowników. Zebrany materiał opisuje trening na danych publicznych, od stron trzecich oraz syntetycznych, ale nie podaje osobnej polityki prywatności ani regionu przechowywania danych dla samego modelu.","2026-07-15",[],"Inkling to pierwszy open-weights model Thinking Machines Lab, ogłoszony 15 lipca 2026. Jest to multimodalny MoE z 975B parametrami totalnymi, 41B aktywnymi, kontekstem do 1M tokenów i natywną obsługą tekstu, obrazu i audio. Producent pozycjonuje go jako bazę do customizacji, fine-tuningu i wdrożeń przez partnerów API oraz self-hosting.","Producent informuje o wcześniejszych ewaluacjach bezpieczeństwa dla zwykłej interakcji człowiek-AI i testów dangerous-capability (CBRN, cyber, loss of control). Wykryte ryzyka resztkowe obejmują m.in. sporadyczną skłonność do odpowiadania na role-play i pośrednio sformułowane szkodliwe prompty; zalecana jest obrona warstwowa po stronie wdrożenia.","Inkling is a 975B-parameter open-weights multimodal AI model from Thinking Machines Lab that supports text, image, and audio input, offers a 1M token context window, and can be fine-tuned via Tinker or used through API partners.",[751,752,753,367,716,754,755,756],"Web","self-hosted Linux\u002FGPU environments","partner inference platforms","llama.cpp","Ollama","LM Studio",[758,759,760,761,762,763,764],"developers","researchers","ML engineers","teams building agentic systems","organizations customizing foundation models","enterprise teams","AI engineers",[766,767,768,769,770,771,772,773,774,613,775,776,777,778,779,780],"agentic coding","tool use","chatbots","retrieval-augmented generation","instruction following","general conversational use","audio transcription and question answering","vision reasoning","fine-tuning\u002Fcustomization for domain-specific workflows","content generation","SEO","automation","transcription","code generation","infrastructure","Thinking Machines Lab","published","2026-08-12T18:20:31.057Z",[733,785,786,787,788],"DeepSeek V3","Nemotron 3 Ultra","GLM 5.2","Kimi K2.5",{"checked_at":783,"ready":429,"failed":790},[],"Nieznana kategoria: infrastruktura\u002Fapi-modele (auto-fix)","```markdown\n# **Inkling – multimodalny model AI do przetwarzania tekstu, obrazów i audio**\n\n**Inkling** to zaawansowany, otwarty model AI o architekturze *Mixture-of-Experts* (MoE) z **975 miliardami parametrów**, opracowany przez *Thinking Machines Lab*. Narzędzie wyróżnia się **natywną multimodalnością** – przetwarza tekst, obrazy i audio (wyjściem jest tekst), a także oferuje **okno kontekstu do 1 mln tokenów**. Jest skierowane do deweloperów, badaczy i firm poszukujących elastycznego, wysokowydajnego modelu do zadań takich jak *agentic coding*, obsługa narzędzi czy tworzenie zaawansowanych chatbotów.\n\n---\n\n## **Najważniejsze funkcje**\n\n1. **Multimodalność w jednym modelu**\n   Inkling akceptuje **tekst, obrazy i audio** jako dane wejściowe, generując tekstowe odpowiedzi. Dzięki temu sprawdza się w zadaniach wymagających analizy wielu typów danych jednocześnie – np. transkrypcja mowy z kontekstem wizualnym czy generowanie opisu obrazu z uwzględnieniem pytań użytkownika.\n\n2. **Ogromne okno kontekstu (1M tokenów)**\n   Model radzi sobie z **długimi dokumentami, kodem źródłowym czy rozmowami**, zachowując spójność nawet przy bardzo rozbudowanych wejściach. To kluczowe dla zastosowań takich jak analiza kodu czy tworzenie agentów AI z pamięcią kontekstową.\n\n3. **Kontrola nad \"wysiłkiem myślowym\" (*thinking effort*)**\n   Użytkownik może **balansować między jakością odpowiedzi a kosztem obliczeniowym**, dostosowując poziom \"zaangażowania\" modelu. Przydatne w scenariuszach, gdzie liczy się szybkość (np. chatboty) lub precyzja (np. analiza danych).\n\n4. **Architektura *Mixture-of-Experts* (MoE)**\n   Inkling wykorzystuje **41 miliardów aktywnych parametrów** (z puli 975 mld), co optymalizuje zużycie zasobów przy zachowaniu wysokiej wydajności. Model dynamicznie aktywuje specjalistyczne podsieci (*experts*) w zależności od zadania.\n\n5. **Fine-tuning i integracje**\n   - Możliwość **dostrajania modelu** za pomocą platformy **Tinker** (z 50% zniżką w okresie promocyjnym).\n   - Dostęp przez **API partnerów** lub pobranie wag modelu z **Hugging Face** (licencja Apache 2.0).\n\n---\n\n## **Dla kogo jest Inkling?**\n\n- **Deweloperzy i inżynierowie AI**\n  Idealny do budowy zaawansowanych agentów, narzędzi do analizy kodu (*agentic coding*) czy systemów automatyzujących workflow z użyciem wielu modalności.\n\n- **Badacze i naukowcy**\n  Otwarty charakter modelu (wagi dostępne na Hugging Face) umożliwia eksperymenty z fine-tuningiem, benchmarkingiem czy badaniami nad multimodalnością.\n\n- **Firmy tworzące chatboty i asystentów głosowych**\n  Dzięki obsłudze audio i tekstu, Inkling może być używany do budowy **multimodalnych asystentów** (np. przetwarzanie pytań głosowych z jednoczesną analizą przesłanych obrazów).\n\n- **Startupy i zespoły R&D**\n  Przystępność (darmowe wagi modelu) i skalowalność (MoE) czynią go atrakcyjnym dla projektów wymagających wysokiej wydajności przy ograniczonych budżetach.\n\n- **Twórcy narzędzi do analizy danych**\n  Przydatny w scenariuszach, gdzie konieczne jest łączenie informacji z różnych źródeł (np. generowanie raportów na podstawie obrazów, tekstu i nagrań).\n\n---\n\n## **Cena**\n\n- **Darmowy plan**:\n  Wagi modelu są **dostępne bezpłatnie** na [Hugging Face](https:\u002F\u002Fhuggingface.co\u002F) pod licencją **Apache 2.0**, co pozwala na lokalne uruchomienie i modyfikacje.\n\n- **Płatne opcje**:\n  - **Fine-tuning na platformie Tinker**: Obecnie oferowany z **50% zniżką** (ceny docelowe nie są jawnie podane – szczegóły mają znaleźć się w dokumentacji).\n  - **API partnerów**: Koszty zależą od dostawcy (brak oficjalnych stawek od Thinking Machines Lab).\n\n*Uwaga*: Producent nie publikuje szczegółowych planów cenowych, co może utrudniać oszacowanie kosztów dla firm planujących komercyjne wdrożenia.\n\n---\n\n## **Zalety i wady**\n\n✅ **Zalety**:\n1. **Otwartość i elastyczność**:\n   Darmowe wagi modelu i licencja Apache 2.0 umożliwiają swobodne eksperymenty i wdrożenia lokalne bez ograniczeń licencyjnych.\n\n2. **Multimodalność w praktyce**:\n   Rzadko spotykana kombinacja obsługi tekstu, obrazów i audio w jednym modelu otwiera nowe możliwości dla aplikacji (np. asystenci rozumiejący kontekst wizualny i głosowy).\n\n3. **Skalowalność dzięki MoE**:\n   Architektura *Mixture-of-Experts* pozwala na efektywne wykorzystanie zasobów, aktywując tylko niezbędne części modelu.\n\n❌ **Wady**:\n1. **Brak transparentnych cen**:\n   Niejasne koszty fine-tuningu i API mogą zniechęcać firmy planujące komercyjne użycie.\n\n2. **Wymagania sprzętowe**:\n   Uruchomienie modelu o 975B parametrach lokalnie wymaga **wysokowydajnej infrastruktury** (np. GPU z dużą pamięcią), co może być barierą dla mniejszych zespołów.\n\n3. **Ograniczone wsparcie dla początkujących**:\n   Brak gotowych, przyjaznych",{"source_page_url":304,"final_url":304,"fetched_at":424},{"alternatives_count":184,"active_deals":184},[],[],{"data":798,"_meta":1704,"alternatives":1705,"deals":1706},{"source":302,"source_key":303,"source_page_url":799,"name":800,"slug":801,"description":802,"title":803,"category":309,"access_model":804,"pricing_evidence":805,"capability_evidence":809,"headings":813,"breadcrumbs":814,"pros":815,"cons":828,"external_links":834,"affiliate_links":879,"official_link_candidates":880,"text_excerpt":891,"fetched_at":892,"final_url":799,"http_status":425,"content_type":426,"content_hash":893,"cache_file":894,"published":429,"verification_status":430,"official_url":844,"official_evidence":895,"api_available":1507,"blog_titles":1508,"data_region":680,"data_training":680,"entity_type":1509,"external_rating":680,"features":1510,"free_summary":1533,"gdpr":680,"integrations":1534,"languages":1537,"license":680,"limitations":1552,"model_names":1558,"polish_support":734,"pricing_plans":1625,"pricing_summary":1644,"privacy_summary":1645,"release_date":1646,"research_links":1647,"research_summary":1648,"security_summary":1649,"self_hosted":1650,"short_description":1651,"starting_price":1628,"supported_platforms":1652,"target_users":1654,"trial_summary":1669,"use_cases":1670,"vendor":1685,"status":782,"last_verified_at":1686,"harvest_verification":1687,"last_harvest_verified_at":1688,"alternatives":1695,"quality":1700,"category_v3":265,"category_v3_reason":791,"deep_card":1702,"deep_card_generated":429,"category_v3_main":259,"category_v3_label":266,"category_v3_main_label":260,"_category_fixed":429,"evidence":1703},"https:\u002F\u002Fwww.aixploria.com\u002Fflaq-ai\u002F","Flaq AI","flaq-ai","Flaq réunit des modèles texte, image et vidéo dans une API unique avec outils gratuits, accès stable et facturation centralisée.","Flaq AI : Avis, Prix, Info & 39 IA Alternatives | 2026 | AIxploria","usage_based",[806,807,808],"Freemium","98 Visiter Freemium Kling 3.0","181 Visiter Freemium Seedance 2.0",[810,322,811,812],"« Accédez en un seul point à des API pour la génération d'images, de vidéos et de texte, avec des modèles comme GPT Image 2, Seedance, Nano Banana, Kimi K3 ou encore Claude Opus 4.7 »","« Le générateur de vidéos IA associé au puissant modèle Grok (xAI). Transformez rapidement vos images en courtes vidéos (disponible sur iOS) »","« Générez des scènes vidéo réalistes en 24 FPS, avec une résolution de 768 x 512 pixels en temps réel à partir de texte ou d'images. Contrôlez les personnages, les décors ainsi que le style grâce à un modèle vidéo multimodal et open source »",[],[],[816,817,818,819,820,821,822,823,824,825,826,827],"Jedno API do wielu klas modeli","Szeroki katalog modeli image\u002Fvideo\u002FLLM","Widoczne API Docs i przykłady integracji","Free tools do szybkiego testowania","Modele i ceny prezentowane na stronie, co ułatwia porównanie","Wsparcie workflow dla twórców, zespołów produktowych i deweloperów","Unified API for multiple AI models (text, image, video)","Cost-effective pricing compared to official APIs","Support for high-resolution 4K image editing","Fast and stable video generation with Veo 3.1 Fast API","Open-source and multimodal video model support","Free tools available for creative workflows",[829,830,831,832,833],"Brak zweryfikowanych publicznie danych o polityce prywatności i treningu danych poza odnośnikiem do polityki\u002Fwarunków na stronie","Nie udało się potwierdzić regionu przechowywania danych ani jednoznacznej zgodności GDPR\u002FRODO","Cennik na stronie pokazuje ceny w kredytach\u002Fna użycie, ale nie ma pełnej, publicznej tabeli wszystkich planów abonamentowych dla całej platformy","Część informacji o modelach i cenach jest snapshotem i może się zmieniać","Brak zweryfikowanej, niezależnej oceny zewnętrznej",[835,837,839,841,843,846,848,850,853,856,859,862,863,864,865,866,867,868,869,870,871,872,873,874,875,876,877,878],{"url":836,"text":352},"https:\u002F\u002Ftwitter.com\u002Fintent\u002Ftweet?url=https%3A%2F%2Fwww.aixploria.com%2Fflaq-ai%2F&text=Flaq+AI%3A+Acc%C3%A9dez+en+un+seul+point+%C3%A0+des+API+pour+la+g%C3%A9n%C3%A9ration+d%27images%2C+de+vid%C3%A9os+et+de+texte%2C+avec+des+mod%C3%A8les+comme+GPT+Image+2%2C+Seedance%2C+Nano+Banana%2C+Kimi+K3+ou+encore+Claude+Opus+4.7",{"url":838,"text":352},"https:\u002F\u002Fwww.facebook.com\u002Fsharer\u002Fsharer.php?u=https%3A%2F%2Fwww.aixploria.com%2Fflaq-ai%2F",{"url":840,"text":352},"https:\u002F\u002Fwww.linkedin.com\u002FshareArticle?url=https%3A%2F%2Fwww.aixploria.com%2Fflaq-ai%2F&title=Flaq+AI",{"url":842,"text":352},"https:\u002F\u002Ft.me\u002Fshare\u002Furl?url=https%3A%2F%2Fwww.aixploria.com%2Fflaq-ai%2F&text=Acc%C3%A9dez+en+un+seul+point+%C3%A0+des+API+pour+la+g%C3%A9n%C3%A9ration+d%27images%2C+de+vid%C3%A9os+et+de+texte%2C+avec+des+mod%C3%A8les+comme+GPT+Image+2%2C+Seedance%2C+Nano+Banana%2C+Kimi+K3+ou+encore+Claude+Opus+4.7",{"url":844,"text":845},"https:\u002F\u002Fflaq.ai\u002F","Visiter ce site",{"url":847,"text":361},"https:\u002F\u002Fchatgpt.com\u002F?q=Donne-moi%20une%20pr%C3%A9sentation%20claire%20et%20concise%20de%20Flaq%20AI.%20Explique%20%C3%A0%20quoi%20sert%20cet%20outil%20IA%2C%20%C3%A0%20qui%20il%20s%E2%80%99adresse%2C%20ses%20principales%20fonctionnalit%C3%A9s%2C%20son%20mod%C3%A8le%20tarifaire%2C%20ses%20points%20forts%2C%20ses%20limites%20et%20ses%20alternatives%20possibles.",{"url":849,"text":364},"https:\u002F\u002Fwww.perplexity.ai\u002F?q=Recherche%20des%20informations%20%C3%A0%20jour%20sur%20Flaq%20AI.%20R%C3%A9sume%20%C3%A0%20quoi%20sert%20cet%20outil%2C%20ses%20prix%20actuels%20si%20disponibles%2C%20ses%20principales%20fonctionnalit%C3%A9s%2C%20ses%20points%20forts%2C%20ses%20limites%20et%20ses%20alternatives%20notables.",{"url":851,"text":852},"https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=RgPcu25Otzc","YouTube",{"url":854,"text":855},"https:\u002F\u002Fx.com\u002Faiflaq","X",{"url":857,"text":858},"https:\u002F\u002Fdiscord.gg\u002FqZwNtm8gnM","Discord",{"url":860,"text":861},"https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fflaqai\u002F","Reddit",{"url":378,"text":379},{"url":378,"text":381},{"url":378,"text":383},{"url":385,"text":386},{"url":385,"text":388},{"url":385,"text":383},{"url":391,"text":392},{"url":391,"text":322},{"url":391,"text":383},{"url":396,"text":397},{"url":396,"text":399},{"url":396,"text":383},{"url":402,"text":352},{"url":404,"text":352},{"url":406,"text":352},{"url":408,"text":409},{"url":408,"text":409},[],[881,882,883,884,885,886,887,888,889,890],{"url":842,"text":352},{"url":844,"text":845},{"url":847,"text":361},{"url":849,"text":364},{"url":860,"text":861},{"url":378,"text":379},{"url":378,"text":381},{"url":378,"text":383},{"url":385,"text":386},{"url":385,"text":388},"Flaq AI : Avis, Prix, Info & 39 IA Alternatives | 2026 | AIxploria \nTop 10 Catégories IA Tutoriels IA + Plus Nouveaux Outils IA Outils IA Gratuits Actualités IA 2026 Bonus & Extras Soumettre une IA \nEnglish Français Español Deutsch Italiano Português + More × 中文 日本語 العربية हिन्दी Русский 한국어 Türkçe Tiếng Việt فارسی বাংলা Indonesia ไทย Nederlands Polski اردو Kiswahili Українська Ελληνικά עברית Svenska Norsk Dansk Suomi Čeština Magyar Melayu Tagalog 中文 (香港) தமிழ் \nS'inscrire \nTop 10 Catégories IA Tutoriels IA + Plus Nouveaux Outils IA Outils IA Gratuits Actualités IA 2026 Bonus & Extras Soumettre une IA \nS'inscrire \nAccueil Générateurs Images Flaq AI \nFlaq AI\n#98 dans Générateurs Vidéos \n4.5\u002F5 \nVisiter ce site \n« Accédez en un seul point à des API pour la génération d'images, de vidéos et de texte, avec des modèles comme GPT Image 2, Seedance, Nano Banana, Kimi K3 ou encore Claude Opus 4.7 »\nEn savoir plus : \nChatGPT \nPerplexity \nFreemium \n5862 \nLiens utiles \nSite Officiel \nYouTube \nX \nDiscord \nReddit \n#Générateurs Images \n#Générateurs Vidéos \n#IA Récentes \n#Outils Développeur C'est un nouvel outil IA \n54 ★ Outils IA de Prestige ★ \n★ Mis en avant \nEmergent AI \n« Transformez vos idées en applications complètes grâce à des agents IA qui codent, testent et déploient pour vous. Créez sites web (full-stack), apps IA, jeux et outils automatisés simplement en décrivant votre projet »\nVisiter \n★ Mis en avant \nClickUp \n« Gérez vos projets plus vite avec une plateforme qui regroupe équipes, tâches et documents en un seul espace. Son IA intégrée résume vos notes, rédige pour vous et organise automatiquement vos priorités »\nVisiter \n★ Mis en avant \nThumbnailCreator.com \n« Créez rapidement des miniatures accrocheuses pour vos vidéos sur YouTube, Twitch et Facebook. Attirez plus de vues et améliorez facilement le taux de clics sur vos clip vidéos »\nVisiter \n★ Mis en avant \nVmake Video Enhancer \n« Améliorez vos vidéos floues et de mauvaise qualité pour les transformer en HD et 4K grâce à l'IA. Vmake accentue les détails de vos vidéos, améliore leur netteté et leur donne un aspect plus professionnel en quelques secondes »\nVisiter \nIA Alternatives à \nFlaq AI \nEssai-Gratuit Deevid AI \n4.5\u002F5 \n« Créez des vidéos de haute qualité à partir de textes, d'images ou de vidéos à l'aide d'une IA intuitive. Transformez en moins d'une minute vos prompts en vidéos réalistes (aucune compétence n'est requise) »\n75 \nVisiter Payant Veo 3.1 \n4.6\u002F5 \n« Générez des vidéos cinématiques avec un son natif synchronisé et un contrôle narratif plus précis. Créez des clips réalistes de 6 secondes pouvant être prolongés jusqu'à 1 minute. Possibilité d'insérer\u002Fsupprimer des éléments et d'obtenir des transitions fluides via Flow »\n98 Visiter Freemium Kling 3.0 \n4.5\u002F5 \n« Générez des vidéos 1080p à 30 images par seconde d'une durée maximale de 15 secondes à partir de texte ou d'images. Profitez de personnages consistants, de mouvements de caméra cinématiques et d'une physique 3D réaliste »\n48 Visiter Payant Gemini Omni \n4.4\u002F5 \n« Générez, modifiez et prolongez des vidéos à partir de simples instructions texte ou multimodales, en utilisant le modèle vidéo le plus avancé de Google intégré à la suite Gemini »\n198 Visiter Payant Grok Imagine \n4.5\u002F5 \n« Le générateur de vidéos IA associé au puissant modèle Grok (xAI). Transformez rapidement vos images en courtes vidéos (disponible sur iOS) »\n181 Visiter Freemium Seedance 2.0 \n4.4\u002F5 \n« Créez des vidéos complètes depuis une simple phrase grâce à l'IA. Vous combinez jusqu'à 12 références (9 images, 3 vidéos, 3 audios de 15s max) pour maintenir la cohérence des personnages. L'outil génère des vidéos de 5 à 12 secondes en 720p-1080p\n»\n282 Visiter Payant Runway Gen-4.5 \n4.7\u002F5 \n« Créez des vidéos cinématiques et ultra-réalistes avec une précision physique, une cohérence temporelle et un contrôle créatif total. Classé parmi les meilleurs modèles vidéo au monde avec 1247 points Elo. Disponible pour tous les styles : photoréalisme, animation, etc. »\n70 Visiter Gratuit LTX-2 \n4.4\u002F5 \n« Générez des scènes vidéo réalistes en 24 FPS, avec une résolution de 768 x 512 pixels en temps réel à partir de texte ou d'images. Contrôlez les personnages, les décors ainsi que le style grâce à un modèle vidéo multimodal et open source »\n54 Visiter \n+ Voir plus d'IA \nCopier le code \nL'intelligence artificielle pour tous\nRessources Tutoriels, astuces et blog Agenda des conférences sur l'IA Glossaire des IA Emplois dans l'IA Newsletter Services pratiques Chaînes YouTube sur l'IA Top 100 IA Liste GPTs Outils IA Hubspot Meilleurs Agents IA Entreprise Ajouter un outil IA Annoncer Mettre à jour votre site Sponsorship ★ Media Kit Qui sommes-nous ? Contactez-nous \n© 2026 AIxploria. Tous droits réservés \nPolitique des Cookies \nConditions d'utilisations Mentions Légales Gérer le consentement aux cookies \nAfin de fournir une expérience optimale, Aixploria utilise des technologies telles que les cookies pour stocker et\u002Fou accéder aux informations relatives à l'appareil. En acceptant ces technologies, vous nous permettez de traiter des données telles que le comportement de navigation ou des identifiants uniques sur ce site. \nFonctionnel \nFonctionnel \nToujours activé \nLe stockage ou l’accès technique est strictement nécessaire dans la finalité d’intérêt légitime de permettre l’utilisation d’un service spécifique explicitement demandé par l’abonné ou l’internaute, ou dans le seul but d’effectuer la transmission d’une communication sur un réseau de communications électroniques. \nPréférences \nPréférences \nLe stockage ou l’accès technique est nécessaire dans la finalité d’intérêt légitime de stocker des préférences qui ne sont pas demandées par l’abonné ou la personne utilisant le service. \nStatistiques \nStatistiques \nLe stockage ou l’accès technique qui est utilisé exclusivement à des fins statistiques. \nLe stockage ou l’accès technique qui est utilisé exclusivement dans des finalités statistiques anonymes. En l’absence d’une assignation à comparaître, d’une conformité volontaire de la part de votre fournisseur d’accès à internet ou d’enregistrements supplémentaires provenant d’une tierce partie, les informations stockées ou extraites à cette seule fin ne peuvent généralement pas être utilisées pour vous identifier. \nMarketing \nMarketing \nLe stockage ou l’accès technique est nécessaire pour créer des profils d’internautes afin d’envoyer des publicités, ou pour suivre l’internaute sur un site web ou sur plusieurs sites web ayant des finalités marketing similaires. \nGérer les options \nGérer les services \nGérer {vendor_count} fournisseurs \nEn savoir plus sur ces finalités \nTout accepter \nTout refuser \nPersonnaliser \nSauvegarder les préférences \nPersonnaliser \n{title} \n{title} \n{title} Gérer le consentement aux cookies \nPour offrir les meilleures expériences, nous utilisons des technologies telles que les cookies pour stocker et\u002Fou accéder aux informations des appareils. Le fait de consentir à ces technologies nous permettra de traiter des données telles que le comportement de navigation ou les ID uniques sur ce site. Le fait de ne pas consentir ou de retirer son consentement peut avoir un effet négatif sur certaines caractéristiques et fonctions. \nFonctionnel \nFonctionnel \nToujours activé \nLe stockage ou l’accès technique est strictement nécessaire dans la finalité d’intérêt légitime de permettre l’utilisation d’un service spécifique explicitement demandé par l’abonné ou l’internaute, ou dans le seul but d’effectuer la transmission d’une communication sur un réseau de communications électroniques. \nPréférences \nPréférences \nLe stockage ou l’accès technique est nécessaire dans la finalité d’intérêt légitime de stocker des préférences qui ne sont pas demandées par l’abonné ou la personne utilisant le service. \nStatistiques \nStatistiques \nLe stockage ou l’accès technique qui est utilisé exclusivement à des fins statistiques. \nLe stockage ou l’accès technique qui est utilisé exclusivement dans des finalités statistiques anonymes. En l’absence d’une assignation à comparaître, d’une conformité volontaire de la part de votre fournisseur d’accès à internet ou d’enregistrements supplémentaires provenant d’une tierce partie, les informations stockées ou extraites à cette seule fin ne peuvent généralement pas être utilisées pour vous identifier. \nMarketing \nMarketing \nLe stockage ou l’accès technique est nécessaire pour créer des profils d’internautes afin d’envoyer des publicités, ou pour suivre l’internaute sur un site web ou sur plusieurs sites web ayant des finalités marketing similaires. \nGérer les options \nGérer les services \nGérer {vendor_count} fournisseurs \nEn savoir plus sur ces finalités \nAccepter \nTout refuser \nPersonnaliser \nSauvegarder les préférences \nPersonnaliser \n{title} \n{title} \n{title} Gérer le consentement Gérer le consentement","2026-08-09T00:14:44.930Z","a9a2ad9e4c9c1037b726671ff02520d7765fdd6555d41e7540e59adb0bc77a26","data\u002Fcatalog\u002Fcache\u002Faixploria\u002F85687ba302e0785a1abcf33d0b3a6190d8bee2990e74adcb09e5f27c563a00f4.html",{"url":844,"final_url":844,"status":425,"fetched_at":896,"content_hash":897,"title":898,"description":899,"headings":900,"pricing_evidence":980,"privacy_evidence":1060,"capability_evidence":1063,"text_excerpt":1083,"linked_pages":1084,"research_links":1494,"blog_titles":1504,"use_case_evidence":1505,"feature_evidence":1506},"2026-08-09T01:04:12.608Z","21004b9bec5e1235838764c128fbfc261eaff8af7185a805896592aae27aba38","Flaq AI: Unified & Affordable AI Agent API for Best Models","Flaq AI is a unified & affordable Agent API platform for advanced AI image, video, and text models. Instantly access top models for AI building with one simple API.",[901,902,903,904,905,906,907,908,909,910,911,912,913,914,915,916,917,918,919,920,921,922,923,924,925,926,927,928,929,930,931,932,933,934,935,936,937,938,939,940,941,942,943,944,945,946,947,948,949,950,951,952,953,954,955,956,957,958,959,960,961,962,963,800,964,965,966,967,968,969,970,971,972,973,974,975,976,977,978,979],"Flaq AI: Unified & Affordable AI Agent APIs for Best Models","Flaq AI: Unified AI Agent APIs for Best Models","Production Image APIs with GPT Image 2","Scalable Video APIs with Seedance V2.0","Advanced LLM APIs for Production Workflows","Free AI Tools for Art Creation","Recently Launched AI Model APIs","Qwen","qwen-image-3.0","qwen-image-3.0-edit","Qwen Image 3.0 Pro API","Qwen Image 3.0 Pro API for Image Editing","Qwen Flash Character API","Qwen Plus Character API","Qwen Image Lora Edit API","Qwen Image 2.0 API","Qwen Image 2.0 Edit API","Qwen Image 2.0 Pro API","Qwen Image 2.0 Pro Edit API","Black Forest Labs","flux-3.0-text-to-video","flux-3.0-image-to-video","flux-3.0-start-end-to-video","flux-3.0-video-extend","MiniMax","MiniMax H3 API for Text-to-Video","MiniMax H3 API for Image-to-Video","MiniMax H3 API for Reference-to-Video","Anthropic","Claude Opus 5 API for Text","Claude Opus 5 API for File Analysis","Claude Sonnet 5 Text API","Claude Sonnet 5 File Analysis API","Claude Fable 5 Text API","Claude Fable 5 Web Search API","Claude Fable 5 File Analysis API","Claude Opus 4.8 Text API","Claude Opus 4.8 File Analysis API","DeepSeek","DeepSeek v4 Flash Text API","DeepSeek v4 Flash Web Search API","DeepSeek v4 Pro Text API","DeepSeek v4 Pro Web Search API","Alibaba","Qwen 3.8 Max API for Text","Qwen 3.8 Max API for Web Search","Wan","Wan 2.7 API for Reference-to-Video","Wan 2.7 Image API","Wan 2.7 Image Edit API","Wan 2.7 Image Pro API","Wan 2.7 Image Pro Edit API","Wan 2.7 Lora API for Image-to-Video","Wan 2.7 API for Text-to-Video","Wan 2.7 API for Image-to-Video","Wan 2.7 API for Video Editing","Vidu","Vidu Q3 API for Reference-to-Video","Kimi","Kimi K3 Text API","Grok","Grok 4.5 Text-to-Text API","Grok 4.5 Image-to-Text API","Video Upscaler API","Video Upscaler Pro API","Seedream","Seedream 5.0 Pro API","Seedream 5.0 Pro Edit API","OpenAI","gpt-5.6-sol-text-to-text","gpt-5.6-sol-image-to-text","gpt-5.6-sol-file-analysis","gpt-5.6-sol-web-search","gpt-5.6-terra-text-to-text","gpt-5.6-terra-image-to-text","gpt-5.6-terra-file-analysis","gpt-5.6-terra-web-search","gpt-5.6-luna-text-to-text","gpt-5.6-luna-image-to-text",[981,906,982,983,984,985,986,987,988,989,990,991,992,993,994,995,996,997,998,999,1000,1001,1002,1003,1004,1005,1006,1007,1008,1009,1010,1011,1012,1013,1014,1015,1016,1017,1018,1019,1020,1021,1022,1023,1024,1025,1026,1027,1028,1029,1030,1031,1032,1033,1034,1035,1036,1037,1038,1039,1040,1041,1042,1043,1044,1045,1046,1047,1048,1049,1050,1051,1052,1053,1054,1055,1056,1057,1058,1059],"Flaq AI: Unified & Affordable AI Agent API for Best Models Image API Video API LLM API AI Tools Free Tools Billing Resources menu Image API Image API Video API Video API LLM API LLM API AI Tools AI Tools Free Tools Free Tools Billing Billing Resources Resources Flaq AI: Unified & Affordable AI Agent APIs for Best Models","Explore Flaq AI's free creative tools for quick image generation, editing, and browser-based workflows. Generate images from text prompts, or upload reference images to guide edits, redesigns, style tests, product concepts, posters, social media graphics, and realistic creative scenes.","Discover the latest AI model releases, compare entry prices, and catch available discounts for your next creative workflow.","From $0.0300 \u002F Image qwen-image-3.0-edit","From $0.0300 \u002F Image Qwen Image 3.0 Pro API","From $0.0350 \u002F Image Qwen Image 3.0 Pro API for Image Editing","From $0.0350 \u002F Image Qwen Flash Character API","From $0.0500 \u002F 1M Tokens Qwen Plus Character API","From $0.5000 \u002F 1M Tokens Qwen Image Lora Edit API","From $0.0400 \u002F Image Qwen Image 2.0 API","From $0.0270 \u002F Image - 10 % Qwen Image 2.0 Edit API","From $0.0270 \u002F Image - 10 % Qwen Image 2.0 Pro API","From $0.0630 \u002F Image - 10 % Qwen Image 2.0 Pro Edit API","From $0.0630 \u002F Image - 10 % Black Forest Labs","From $0.1700 \u002F Second flux-3.0-image-to-video","From $0.1700 \u002F Second flux-3.0-start-end-to-video","From $0.1700 \u002F Second flux-3.0-video-extend","From $0.1700 \u002F Second MiniMax","From $0.0900 \u002F Second MiniMax H3 API for Image-to-Video","From $0.0900 \u002F Second MiniMax H3 API for Reference-to-Video","From $0.0900 \u002F Second Anthropic","From $4.5000 \u002F 1M Tokens - 10 % Claude Opus 5 API for File Analysis","From $4.5000 \u002F 1M Tokens - 10 % Claude Sonnet 5 Text API","From $1.8000 \u002F 1M Tokens - 10 % Claude Sonnet 5 File Analysis API","From $1.8000 \u002F 1M Tokens - 10 % Claude Fable 5 Text API","From $9.0000 \u002F 1M Tokens - 10 % Claude Fable 5 Web Search API","From $9.0000 \u002F 1M Tokens - 10 % Claude Fable 5 File Analysis API","From $9.0000 \u002F 1M Tokens - 10 % Claude Opus 4.8 Text API","From $4.5000 \u002F 1M Tokens - 10 % Claude Opus 4.8 File Analysis API","From $4.5000 \u002F 1M Tokens - 10 % DeepSeek","From $0.1260 \u002F 1M Tokens - 10 % DeepSeek v4 Flash Web Search API","From $0.1260 \u002F 1M Tokens - 10 % DeepSeek v4 Pro Text API","From $2.1600 \u002F 1M Tokens - 10 % DeepSeek v4 Pro Web Search API","From $2.1600 \u002F 1M Tokens - 10 % Alibaba","From $2.0000 \u002F 1M Tokens Qwen 3.8 Max API for Web Search","From $2.0000 \u002F 1M Tokens Wan","From $0.1350 \u002F Second - 10 % Wan 2.7 Image API","From $0.0270 \u002F Image - 10 % Wan 2.7 Image Edit API","From $0.0270 \u002F Image - 10 % Wan 2.7 Image Pro API","From $0.0675 \u002F Image - 10 % Wan 2.7 Image Pro Edit API","From $0.0675 \u002F Image - 10 % Wan 2.7 Lora API for Image-to-Video","From $0.1300 \u002F Second Wan 2.7 API for Text-to-Video","From $0.0900 \u002F Second - 10 % Wan 2.7 API for Image-to-Video","From $0.0900 \u002F Second - 10 % Wan 2.7 API for Video Editing","From $0.0900 \u002F Second - 10 % Vidu","From $0.0700 \u002F Second Kimi","From $2.8500 \u002F 1M Tokens - 5 % Grok","From $1.8000 \u002F 1M Tokens - 10 % Grok 4.5 Image-to-Text API","From $1.8000 \u002F 1M Tokens - 10 % Flaq AI","From $0.0050 \u002F Second Video Upscaler Pro API","From $0.0200 \u002F Second Seedream","From $0.0427 \u002F Image - 5 % Seedream 5.0 Pro Edit API","From $0.0427 \u002F Image - 5 % OpenAI","From $5.0000 \u002F 1M Tokens gpt-5.6-sol-image-to-text","From $5.0000 \u002F 1M Tokens gpt-5.6-sol-file-analysis","From $5.0000 \u002F 1M Tokens gpt-5.6-sol-web-search","From $5.0000 \u002F 1M Tokens gpt-5.6-terra-text-to-text","From $2.5000 \u002F 1M Tokens gpt-5.6-terra-image-to-text","From $2.5000 \u002F 1M Tokens gpt-5.6-terra-file-analysis","From $2.5000 \u002F 1M Tokens gpt-5.6-terra-web-search","From $2.5000 \u002F 1M Tokens gpt-5.6-luna-text-to-text","From $1.0000 \u002F 1M Tokens gpt-5.6-luna-image-to-text","From $1.0000 \u002F 1M Tokens gpt-5.6-luna-file-analysis","From $1.0000 \u002F 1M Tokens gpt-5.6-luna-web-search","From $1.0000 \u002F 1M Tokens GPT Image 2 API","From $0.0080 \u002F Image - 20 % GPT Image 2 Edit API","From $0.0240 \u002F Image - 20 % GPT 5.5 Text API","From $5.0000 \u002F 1M Tokens GPT 5.5 Web Search API","From $5.0000 \u002F 1M Tokens GPT 5.5 File Analysis API","From $5.0000 \u002F 1M Tokens Seedance","From $0.0540 \u002F Second - 10 % Seedance V2.0 Mini Image-to-Video API","From $0.0540 \u002F Second - 10 % Seedance V2.0 Text-to-Video API","From $0.0960 \u002F Second - 20 % Seedance V2.0 Image-to-Video API","From $0.0960 \u002F Second - 20 % Seedance V2.0 Fast Text-to-Video API","From $0.0800 \u002F Second - 20 % Seedance V2.0 Fast Image-to-Video API","From $0.0800 \u002F Second - 20 % Happy Horse","From $0.1120 \u002F Second - 20 % Happy Horse 1.1 Image-to-Video API","From $0.1120 \u002F Second - 20 % Happy Horse 1.1 Reference-to-Video API","From $0.1120 \u002F Second - 20 % Happy Horse 1.0 Text-to-Video API",[1061,1062],"How does the platform ensure data security and privacy?","About Us Privacy policy Terms & Conditions Affiliate Agreement Refund Policy Documentation Contact Us Business Flaq AI",[981,1064,902,1065,1066,904,905,1067,1068,985,986,987,988,989,990,991,992,993,926,999,1000,930,1002,1003,1004,1005,1006,1007,1008,1009,940,1011,1012,1013,945,1015,948,1017,1018,1019,1020,1021,1022,1023,1024,958,960,962,1028,964,1030,967,1032,1045,1046,1047,1048,1049,1069,1051,1052,1053,1054,1055,1070,1057,1058,1059,1071,1072,1073,1074,1075,1076,1077,1078,1079,1080,1081,1082],"Try the world’s best video, image, music, and LLM models through one API, with better rates and cost-effectiveness than official APIs.","Try the world’s top video, image, music, and LLM models through a single API, with better rates and cost-effectiveness than official APIs.","Try Image API Try Video API Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Production Image APIs with GPT Image 2","Build reasoning, coding, writing, automation, and file analysis workflows with Claude Opus 4.7, GPT 5.5, and other leading LLM APIs through Flaq AI's unified access layer.","Video AI Video AI Image AI Reset 0 \u002F 2000 480P 21:9 4s Image to Video seedance-v2.0-image-to-video Show labels Panel Inspire me Generate Loading... Recently Launched AI Model APIs","Seedance V2.0 Mini Text-to-Video API","Happy Horse 1.1 Text-to-Video API","From $0.1120 \u002F Second - 20 % Happy Horse 1.0 Image-to-Video API","From $0.0960 \u002F Second - 20 % Kling 3.0 Standard Image-to-Video API","From $0.0840 \u002F Run Kling 3.0 Standard Text-to-Video API","From $0.0840 \u002F Run Kling 3.0 Pro Image-to-Video API","From $0.1120 \u002F Run Kling 3.0 Pro Text-to-Video API","Nano Banana Pro API","From $0.1120 \u002F Image - 20 % Nano Banana Pro Edit API","From $0.1120 \u002F Image - 20 % Nano Banana 2 API","From $0.0560 \u002F Image - 20 % Nano Banana 2 Edit API","View More Models OpenAI Image To Image GPT Image 2 Edit API","GPT Image 2 Edit API provides OpenAI image editing with prompt-guided changes, strong instruction following, detailed visual control, and reliable output for creative production workflows.","From $0.0240 \u002F Image - 20 % Google Image To Image Nano Banana Pro Edit API","Flaq AI: Unified & Affordable AI Agent API for Best Models Image API Video API LLM API AI Tools Free Tools Billing Resources menu Image API Image API Video API Video API LLM API LLM API AI Tools AI Tools Free Tools Free Tools Billing Billing Resources Resources Flaq AI: Unified & Affordable AI Agent APIs for Best Models\nTry the world’s best video, image, music, and LLM models through one API, with better rates and cost-effectiveness than official APIs.\nFlaq AI: Unified AI Agent APIs for Best Models\nTry the world’s top video, image, music, and LLM models through a single API, with better rates and cost-effectiveness than official APIs.\nTry Image API Try Video API Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Claude Claude GPT GPT Hermes Hermes Grok Grok HappyHorse HappyHorse Kling Kling Nano Banana Nano Banana DeepSeek DeepSeek Pixverse Pixverse Seedance Seedance Seedream Seedream Veo Veo Vidu Vidu Wan Wan Kimi Kimi GLM GLM FLUX FLUX MiniMax MiniMax Production Image APIs with GPT Image 2\nCreate polished product shots, marketing visuals, and branded assets with GPT Image 2. It supports strong instruction following, clean text rendering, and fast visual iteration through Flaq AI.\nScalable Video APIs with Seedance V2.0\nTurn product ideas, storyboards, and reference images into dynamic AI videos with Seedance V2.0. It is built for short-form creative workflows, optional sound, and scalable video production.\nAdvanced LLM APIs for Production Workflows\nBuild reasoning, coding, writing, automation, and file analysis workflows with Claude Opus 4.7, GPT 5.5, and other leading LLM APIs through Flaq AI's unified access layer.\nFree AI Tools for Art Creation\nExplore Flaq AI's free creative tools for quick image generation, editing, and browser-based workflows. Generate images from text prompts, or upload reference images to guide edits, redesigns, style tests, product concepts, posters, social media graphics, and realistic creative scenes.\nVideo AI Video AI Image AI Reset 0 \u002F 2000 480P 21:9 4s Image to Video seedance-v2.0-image-to-video Show labels Panel Inspire me Generate Loading... Recently Launched AI Model APIs\nDiscover the latest AI model releases, compare entry prices, and catch available discounts for your next creative workflow.\nQwen\nqwen-image-3.0\nText To Image\nFrom $0.0300 \u002F Image qwen-image-3.0-edit\nImage To Image\nFrom $0.0300 \u002F Image Qwen Image 3.0 Pro API\nText To Image\nFrom $0.0350 \u002F Image Qwen Image 3.0 Pro API for Image Editing\nImage To Image\nFrom $0.0350 \u002F Image Qwen Flash Character API\nText To Text\nFrom $0.0500 \u002F 1M Tokens Qwen Plus Character API\nText To Text\nFrom $0.5000 \u002F 1M Tokens Qwen Image Lora Edit API\nImage To Image\nFrom $0.0400 \u002F Image Qwen Image 2.0 API\nText To Image\nFrom $0.0270 \u002F Image - 10 % Qwen Image 2.0 Edit API\nImage To Image\nFrom $0.0270 \u002F Image - 10 % Qwen Image 2.0 Pro API\nText To Image\nFrom $0.0630 \u002F Image - 10 % Qwen Image 2.0 Pro Edit API\nImage To Image\nFrom $0.0630 \u002F Image - 10 % Black Forest Labs\nflux-3.0-text-to-video\nText To Video\nFrom $0.1700 \u002F Second flux-3.0-image-to-video\nImage To Video\nFrom $0.1700 \u002F Second flux-3.0-start-end-to-video\nImage To Video\nFrom $0.1700 \u002F Second flux-3.0-video-extend\nVideo Editing\nFrom $0.1700 \u002F Second MiniMax\nMiniMax H3 API for Text-to-Video\nText To Video\nFrom $0.0900 \u002F Second MiniMax H3 API for Image-to-Video\nImage To Video\nFrom $0.0900 \u002F Second MiniMax H3 API for Reference-to-Video\nReference To Video\nFrom $0.0900 \u002F Second Anthropic\nClaude Opus 5 API for Text\nText To Text\nFrom $4.5000 \u002F 1M Tokens - 10 % Claude Opus 5 API for File Analysis\nFile Analysis\nFrom $4.5000 \u002F 1M Tokens - 10 % Claude Sonnet 5 Text API\nText To Text\nFrom $1.8000 \u002F 1M Tokens - 10 % Claude Sonnet 5 File Analysis API\nFile Analysis\nFrom $1.8000 \u002F 1M Tokens - 10 % Claude Fable 5 Text API\nText To Text\nFrom $9.0000 \u002F 1M Tokens - 10 % Claude Fable 5 Web Search API\nWeb Search\nFrom $9.0000 \u002F 1M Tokens - 10 % Claude Fable 5 File Analysis API\nFile Analysis\nFrom $9.0000 \u002F 1M Tokens - 10 % Claude Opus 4.8 Text API\nText To Text\nFrom $4.5000 \u002F 1M Tokens - 10 % Claude Opus 4.8 File Analysis API\nFile Analysis\nFrom $4.5000 \u002F 1M Tokens - 10 % DeepSeek\nDeepSeek v4 Flash Text API\nText To Text\nFrom $0.1260 \u002F 1M Tokens - 10 % DeepSeek v4 Flash Web Search API\nWeb Search\nFrom $0.1260 \u002F 1M Tokens - 10 % DeepSeek v4 Pro Text API\nText To Text\nFrom $2.1600 \u002F 1M Tokens - 10 % DeepSeek v4 Pro Web Search API\nWeb Search\nFrom $2.1600 \u002F 1M Tokens - 10 % Alibaba\nQwen 3.8 Max API for Text\nText To Text\nFrom $2.0000 \u002F 1M Tokens Qwen 3.8 Max API for Web Search\nWeb Search\nFrom $2.0000 \u002F 1M Tokens Wan\nWan 2.7 API for Reference-to-Video\nReference To Video\nFrom $0.1350 \u002F Second - 10 % Wan 2.7 Image API\nText To Image\nFrom $0.0270 \u002F Image - 10 % Wan 2.7 Image Edit API\nImage To Image\nFrom $0.0270 \u002F Image - 10 % Wan 2.7 Image Pro API\nText To Image\nFrom $0.0675 \u002F Image - 10 % Wan 2.7 Image Pro Edit API\nImage To Image\nFrom $0.0675 \u002F Image - 10 % Wan 2.7 Lora API for Image-to-Video\nImage To Video\nFrom $0.1300 \u002F Second Wan 2.7 API for Text-to-Video\nText To Video\nFrom $0.0900 \u002F Second - 10 % Wan 2.7 API for Image-to-Video\nImage To Video\nFrom $0.0900 \u002F Second - 10 % Wan 2.7 API for Video Editing\nVideo Editing\nFrom $0.0900 \u002F Second - 10 % Vidu\nVidu Q3 API for Reference-to-Video\nReference To Video\nFrom $0.0700 \u002F Second Kimi\nKimi K3 Text API\nText To Text\nFrom $2.8500 \u002F 1M Tokens - 5 % Grok\nGrok 4.5 Text-to-Text API\nText To Text\nFrom $1.8000 \u002F 1M Tokens - 10 % Grok 4.5 Image-to-Text API\nImage To Text\nFrom $1.8000 \u002F 1M Tokens - 10 % Flaq AI\nVideo Upscaler API\nVideo Upscaler\nFrom $0.0050 \u002F Second Video Upscaler Pro API\nVideo Upscaler\nFrom $0.0200 \u002F Second Seedream\nSeedream 5.0 Pro API\nText To Image\nFrom $0.0427 \u002F Image - 5 % Seedream 5.0 Pro Edit API\nImage To Image\nFrom $0.0427 \u002F Image - 5 % OpenAI\ngpt-5.6-sol-text-to-text\nText To Text\nFrom $5.0000 \u002F 1M Tokens gpt-5.6-sol-image-to-text\nImage To Text\nFrom $5.0000 \u002F 1M Tokens gpt-5.6-sol-file-analysis\nFile Analysis\nFrom $5.0000 \u002F 1M Tokens gpt-5.6-sol-web-search\nWeb Search\nFrom $5.0000 \u002F 1M Tokens gpt-5.6-terra-text-to-text\nText To Text\nFrom $2.5000 \u002F 1M Tokens gpt-5.6-terra-image-to-text\nImage To Text\nFrom $2.5000 \u002F 1M Tokens gpt-5.6-terra-file-analysis\nFile Analysis\nFrom $2.5000 \u002F 1M Tokens gpt-5.6-terra-web-search\nWeb Search\nFrom $2.5000 \u002F 1M Tokens gpt-5.6-luna-text-to-text\nText To Text\nFrom $1.0000 \u002F 1M Tokens gpt-5.6-luna-image-to-text\nImage To Text\nFrom $1.0000 \u002F 1M Tokens gpt-5.6-luna-file-analysis\nFile Analysis\nFrom $1.0000 \u002F 1M Tokens gpt-5.6-luna-web-search\nWeb Search\nFrom $1.0000 \u002F 1M Tokens GPT Image 2 API\nText To Image\nFrom $0.0080 \u002F Image - 20 % GPT Image 2 Edit API\nImage To Image\nFrom $0.0240 \u002F Image - 20 % GPT 5.5 Text API\nText To Text\nFrom $5.0000 \u002F 1M Tokens GPT 5.5 Web Search API\nWeb Search\nFrom $5.0000 \u002F 1M Tokens GPT 5.5 File Analysis API\nFile Analysis\nFrom $5.0000 \u002F 1M Tokens Seedance\nSeedance V2.0 Mini Text-to-Video API\nText To Video\nFrom $0.0540 \u002F Second - 10 % Seedance V2.0 Mini Image-to-Video API\nImage To Video\nFrom $0.0540 \u002F Second - 10 % Seedance V2.0 Text-to-Video API\nText To Video\nFrom $0.0960 \u002F Second - 20 % Seedance V2.0 Image-to-Video API\nImage To Video\nFrom $0.0960 \u002F Second - 20 % Seedance V2.0 Fast Text-to-Video API\nText To Video\nFrom $0.0800 \u002F Second - 20 % Seedance V2.0 Fast Image-to-Video API\nImage To Video\nFrom $0.0800 \u002F Second - 20 % Happy Horse\nHappy Horse 1.1 Text-to-Video API\nText To Video\nFrom $0.1120 \u002F Second - 20 % Happy Horse 1.1 Image-to-Video API\nImage To Video\nFrom $0.1120 \u002F Second - 20 % Happy Horse 1.1 Reference-to-Video API\nReference To Video\nFrom $0.1120 \u002F Second - 20 % Happy Horse 1.0 Text-to-Video API\nText To Video\nFrom $0.1120 \u002F Second - 20 % Happy Horse 1.0 Image-to-Video API\nImage To Video\nFrom $0.1120 \u002F Second - 20 % Kling\nkling-3.0-turbo-text-to-video\nText To Video\nFrom $0.0960 \u002F Second - 20 % kling-3.0-turbo-image-to-video\nImage To Video\nFrom $0.0960 \u002F Second - 20 % Kling 3.0 Standard Image-to-Video API\nImage To Video\nFrom $0.0840 \u002F Run Kling 3.0 Standard Text-to-Video API\nText To Video\nFrom $0.0840 \u002F Run Kling 3.0 Pro Image-to-Video API\nImage To Video\nFrom $0.1120 \u002F Run Kling 3.0 Pro Text-to-Video API\nText To Video\nFrom $0.1120 \u002F Run Google\nNano Banana Pro API\nText To Image\nFrom $0.1120 \u002F Image - 20 % Nano Banana Pro Edit API\nImage To Image\nFrom $0.1120 \u002F Image - 20 % Nano Banana 2 API\nText To Image\nFrom $0.0560 \u002F Image - 20 % Nano Banana 2 Edit API\nImage To Image\nFrom $0.0560 \u002F Image - 20 % Recommended AI Models for Media Creation\nExplore representative AI image, video, LLM or other models for production-ready creative workflows, with live pricing and discounts through Flaq AI's unified model market.\nView More Models OpenAI Image To Image GPT Image 2 Edit API\nGPT Image 2 Edit API provides OpenAI image editing with prompt-guided changes, strong instruction following, detailed visual control, and reliable output for creative production workflows.\nFrom $0.0240 \u002F Image - 20 % Google Image To Image Nano Banana Pro Edit API\nNano Banana Pro Edit API provides Gemini 3.0 Pro image editing with precise local modifications, strong prompt understanding, native 4K output, and reliable results for professional workflows.\nFrom $0.1120 \u002F Image - 20 % Google Text To Image Nano Banana 2 API\nNano Banana 2 API offers fast, affordable Gemini image generation with near-Pro quality, strong prompt understanding, and scalable output for users seeking value, speed, and reliable visual creation.\nFrom $0.0560 \u002F Image - 20 % ByteDance Image To Image Seedream V5.0 Edit API\nSeedream V5.0 Edit API focuses on intelligence-first ByteDance image editing with strong prompt adherence, semantic reasoning, and precise visual control for high-quality production workflows.\nFrom $0.0315 \u002F Run - 10 % ByteDance Image To Video Seedance V2.0 Image-to-Video API\nSeedance V2.0 Image-to-Video API offers professional ByteDance image animation with optional sound, fixed camera control, flexible duration, and broad format coverage.\nFrom $0.0960 \u002F Second - 20 % Alibaba Image To Video Wan 2.7 API for Image-to-Video\nWan 2.7 API for image-to-video provides advanced Alibaba image animation with strong prompt adherence, physics-aware motion, optional audio, 1080p output, and professional visual quality.\nFrom $0.0900 \u002F Second - 10 % Kuaishou Image To Video Kling 3.0 Standard Image-to-Video API\nKling 3.0 Standard Image-to-Video API offers cost-effective Kuaishou image animation for scalable production, stable motion, and reliable short-form video generation workflows.\nFrom $0.0840 \u002F Run Google Image To Video Veo 3.1 Fast Image-to-Video API\nVeo 3.1 Fast Image-to-Video API provides rapid Google image animation with cinematic motion, cost-effective output, and scalable performance for high-volume creative workflows.\nFrom $1.1400 \u002F Run - 5 % Anthropic File Analysis Claude Opus 4.8 File Analysis API\nClaude Opus 4.8 File Analysis API supports deep document review, report summarization, extraction, and high-accuracy file reasoning workflows.\nFrom $4.5000 \u002F 1M Tokens - 10 % OpenAI Text To Text GPT 5.5 Text API\nGPT 5.5 Text API provides advanced OpenAI reasoning, writing, coding, and analysis for complex production tasks that need stronger accuracy and depth.\nFrom $5.0000 \u002F 1M Tokens xAI Image To Text Grok 4 Image-to-Text API\nGrok 4 Image-to-Text API provides xAI multimodal understanding for visual Q&A, OCR, image analysis, and production AI applications.\nFrom $3.0000 \u002F 1M Tokens OpenAI Web Search GPT 5.5 Web Search API\nGPT 5.5 Web Search API combines advanced OpenAI reasoning with web search for research, current information, analysis, and grounded production workflows.\nFrom $5.0000 \u002F 1M Tokens Explore Free Productivity Toolset in Flaq AI\nBeyond powerful enterprise-grade API services, Flaq AI is dedicated to lowering the barrier to AI creation—offering personal users and developers a suite of free practical tools built atop our proprietary infrastructure.\nFree Nano Banana AI\nThe platform now offers free access to the Nano Banana model for image generation. Powered by Flaq AI's elastic compute pool, this tool helps users rapidly convert ideas into visual assets.\nMinimal operation: No complex API setup—generate images instantly from your ideas right in your browser.\nInstant feedback: Flaq’s millisecond-level inference optimization delivers smooth, frustration-free generation with no long waits.\nPrototype validation: Generated images are ready for product prototyping or creative exploration, empowering rapid iteration for developers.\nFree GPT Image 2\nFree to create detailed visuals with a free GPT Image 2 generator and AI image editor. Generate images from text prompts, or upload reference images to guide edits, redesigns, style tests, product concepts, posters, social media graphics, and realistic creative scenes.\nFree Image to Image Generator\nFree to transform uploaded visuals with a free Image to Image generator and AI image editor. Use reference images to guide edits, redesigns, style changes, product concepts, and creative variations without signup or a credit card.\nFree Image Editor\nFree to create and edit detailed visuals with a free AI image editor powered by GPT Image 2. Generate images from text prompts, or upload reference images to guide edits, redesigns, style tests, product concepts, posters, social media graphics, and realistic creative scenes.\nExplore multiple Creative Tools with Flaq AI\nExplore several AI creation tools for quick image and video workflows in your browser, then scale successful ideas with Flaq AI's production-ready model APIs. Flaq AI provides a unified API layer for all models, making it easy to use and scale your workflows.\nAI Text to Image Generator\nCreate polished images from prompts with leading AI image models, flexible settings, and a fast browser-based workflow.\nAI Image to Image Generator\nUpload a reference image, guide edits with prompts, and transform visuals for design, marketing, and creative production.\nAI Text to Video Generator\nTurn written scene ideas into short AI videos with model selection, motion prompts, and practical generation controls.\nAI Image to Video Generator\nAnimate reference images into smooth AI video clips for products, portraits, social posts, and creative concepts.\nTypical Application Scenarios in Flaq AI\nFlaq AI's high-performance API cluster is more than just a set of tools—it's the driving engine for industry innovation. From creative workflows to enterprise automation, we provide technical support covering all scenarios.\nTry Now Fully Automated Content Marketing Platform\nBatch-generate branded visuals with the Nano Banana Pro API and automatically create posts with LLMs, achieving instant output from copywriting to visuals.\nReal-Time Interactive AI Products\nPowered by Flaq's ultra-low latency inference, enable seamless image and video generation for social apps, games, or real-time assistants.\nFilm & Advertising Industry Pipeline\nIntegrate Veo 3.1 and Wan 2.6 APIs for rapid pre-visualization (Pre-viz) during early script development, greatly reducing traditional rendering costs.\nDeveloper Prototyping & Growth\nQuickly test various model combinations with a unified API key and switch core algorithms anytime based on market feedback, all without code changes.\nFAQs about Flaq AI\nWhat are the differences between Flaq AI and using model providers' official APIs directly? \nWhat types of models does Flaq AI support? \nHow does the platform ensure data security and privacy? \nIs there a free trial quota available? \nCan I use Flaq AI for commercial purposes? \nHow can I get a refund from flaq.ai? \nWhat should I do if I have product questions or encounter issues? \nMore Articles & News about Flaq AI Generation Platform\n2026-08-05 MiniMax H3 API and Open Source Video Creation vs Seedance 2.0\nThe MiniMax H3 API gives video teams a useful combination: image-to-video generation, 768p and 2K output options, clips up to 15 seconds, and a lower listed cost than several comparable Seedance 2.0 settings. MiniMax has also published H3-Base weights on Hugging Face, creating new options for research, custom pipelines, and self-hosted experimentation where its license permits.\nMiniMax H3 AI Video Seedance 2.0 Hailuo 03 Read More 2026-07-22 Kimi K3 API Explained: What Developers Should Know Before Flaq AI Access Arrives\nExplore Kimi K3 for coding, multi",[1085,1135,1183,1225,1273,1311,1350,1374,1398,1422,1447,1471],{"url":1086,"kind":1087,"anchor":1088,"title":1089,"description":1090,"headings":1091,"feature_evidence":1106,"use_case_evidence":1124,"fetched_at":1133,"status":425,"content_hash":1134},"https:\u002F\u002Fflaq.ai\u002Fmodels\u002Fbytedance\u002Fseedream-5-0-edit\u002F","use_case","ByteDance Image To Image Seedream V5.0 Edit API\nSeedream V5.0 Edit API focuses on intelligence-first ByteDance image editing with strong prompt adherence, semantic reasoning, and p","Stable & Affordable ByteDance Seedream 5.0 Edit 4K API | Stable API for Creative Production","Cost-effective 4K image editing via Seedream 5.0 API. Optimized for high-concurrency stability and superior texture consistency at an affordable rate. Built for free testing and stable API workflows.",[1092,1093,1094,1095,1096,1097,1098,1099,1100,1101,1102,1103,1104,1103,1104,1103,1104,1105],"Related Seedream 5.0 Models","Seedream 5.0 Edit Pricing","Examples","Related Models","README","Fast & Affordable SeeDream V5.0 Edit API (ByteDance 4K Intelligence-First Image Editing Integration)","Key Features of SeeDream V5.0 Edit API","How to Use SeeDream V5.0 Edit API for Professional Image Editing on Flaq AI","Best Use Cases for SeeDream V5.0 Edit API Integration","SeeDream V5.0 Edit vs Competitors: Comparative Analysis","API Examples","Submit Example","Polling Example","More Articles for Seedream 5.0 Edit",[1107,1108,1109,1110,1098,1111,1112,1113,1114,1115,1116,1117,1099,1118,1119,1120,1100,1121,1122,1123],"Stable & Affordable ByteDance Seedream 5.0 Edit 4K API | Stable API for Creative Production Image API Video API LLM API AI Tools Free Tools Billing Resources menu Image API Image API Nano Banana Pro Nano Banana Pro Nano Banana 2 Nano Banana 2 GPT Image 2 GPT Image 2 Seedream 5.0 Pro Seedream 5.0 Pro Seedream 5.0 Seedream 5.0 Seedream 4.5 Seedream 4.5 Grok Imagine Grok Imagine Qwen Image 2.0 Qwen Image 2.0 Wan 2.7 Image Wan 2.7 Image Qwen Image Lora Qwen Image Lora Z Image Z Image Qwen Image 3.0 Pro Qwen Image 3.0 Pro Video API Video API LLM API LLM API AI Tools AI Tools Free Tools Free Tools Billing Billing Resources Resources ByteDance \u002F seedream-v5.0-edit seedream-v5.0-edit Cost-effective 4K image editing via Seedream 5.0 API. Optimized for high-concurrency stability and superior texture consistency at an affordable rate. Built for free testing and stable API workflows.","seedream-v5.0 text-to-image seedream-v5.0-edit image-to-image Documentation Playground API ALL History Documentation Seedream 5.0 Edit Upload Images Click or drag to upload Supported formats : jpg, png 0 \u002F 10 images Enter Prompt Translate copy 0 \u002F 2000 Optimize Prompt ratio resolution Generate Loading... loading... Try the AI Image Generator now","Fast & Affordable SeeDream V5.0 Edit API (ByteDance 4K Intelligence-First Image Editing Integration) Key Features of SeeDream V5.0 Edit API How to Use SeeDream V5.0 Edit API for Professional Image Editing on Flaq AI Best Use Cases for SeeDream V5.0 Edit API Integration SeeDream V5.0 Edit vs Competitors: Comparative Analysis Fast & Affordable SeeDream V5.0 Edit API (ByteDance 4K Intelligence-First Image Editing Integration)","ByteDance SeeDream V5.0 Edit API (powered by the SeeDream V5.0 model) delivers cost-effective, production-grade AI image editing for developers and creative teams. This intelligence-first 4K image editing API integration helps you transform existing visuals into professional outputs through natural-language instructions. The SeeDream V5.0 model provides advanced semantic reasoning and precise editing control, while the API provides stable integration for scalable workflows on Flaq AI.","Native 4K High-Fidelity Image Editing: Produce crisp, production-ready 4K edited images with fine detail, realistic textures, and clean edges through affordable SeeDream image editing API integration.","Intelligence-First Semantic Image Editing: Edit images with natural-language prompts while preserving object relationships, lighting logic, and scene coherence using SeeDream V5.0's advanced reasoning capabilities.","Precise Editing Control: SeeDream V5.0 Edit API delivers example-based controllable editing, enabling targeted modifications with high accuracy—ideal for professional design and production workflows.","Enhanced Prompt Adherence: Leverage significantly improved instruction-following for precise, predictable edits across complex multi-element scenes through cost-effective API integration.","Professional Camera-Style API Controls: Adjust lens perspective, focus, depth of field (bokeh), and color grading for advanced visual direction in API-based editing pipelines.","Flexible Aspect Ratio Support: API supports 7 output formats (16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16), suitable for ads, social media, product pages, and cinematic layouts.","Consistent Character & Brand Rendering: Maintain identity consistency, brand elements, and visual style across multiple edited images through cost-effective API integration.","Input: Existing image(s) (JPEG\u002FPNG) plus natural-language editing prompts","Output: High-resolution 4K edited images (JPEG\u002FPNG) through SeeDream V5.0 Edit API integration","Resolution: Flexible API support for 2K direct output and 4K AI-enhanced resolution; width and height range: 1024–4096","Marketing & Branding: Rapidly localize and refresh campaign creatives with high-quality AI image editing API workflows grounded in intelligence-first semantic understanding.","Product Photography Optimization: Update backgrounds, lighting, and composition for catalog-ready product visuals without full reshoots using affordable API integration.","Social Media Content Production: Edit and create high-volume image variants with consistent style and brand alignment for multi-platform publishing through cost-effective SeeDream API.",[1107,1108,1109,1110,1125,1126,1127,1100,1128,1129,1130,1131,1132],"Precise Editing Control: SeeDream V5.0 Edit API delivers example-based controllable editing, enabling targeted modifications with high accuracy—ideal for professional design and production workflows. ","Output: High-resolution 4K edited images (JPEG\u002FPNG) through SeeDream V5.0 Edit API integration ","Resolution: Flexible API support for 2K direct output and 4K AI-enhanced resolution; width and height range: 1024–4096 ","Marketing & Branding: Rapidly localize and refresh campaign creatives with high-quality AI image editing API workflows grounded in intelligence-first semantic understanding. ","SeeDream 4.5 Edit delivers strong semantic editing with multilingual typography and multi-image batch processing. SeeDream V5.0 Edit advances further with intelligence-first reasoning and enhanced prompt adherence—making it the superior choice for precise, instruction-accurate editing workflows.","FLUX.1 [dev] focuses on detail-heavy technical generation workflows. SeeDream V5.0 Edit API emphasizes semantic, instruction-driven editing with layout-aware changes and precise controllable modifications—making it stronger for practical production editing scenarios using affordable ByteDance integration.","Nano Banana 2 Edit leverages Google's Gemini 3.1 Flash for fast, context-aware editing. SeeDream V5.0 Edit API counters with higher maximum resolution output, intelligence-first semantic reasoning, and precise controllable editing—ideal for applications requiring 4K fidelity and instruction accuracy.","Grok Imagine Edit offers strong multi-style editing with xAI's Aurora engine. SeeDream V5.0 Edit API differentiates through ByteDance's advanced computer vision research, intelligence-first reasoning, and higher resolution output—making it the preferred choice for professional 4K editing workflows.","2026-08-09T01:04:03.026Z","cfa0fcac5e3f154f2c6a258978adf212e8e24cf485946cbae6bcc48ad95ee238",{"url":1136,"kind":1087,"anchor":1137,"title":1138,"description":1139,"headings":1140,"feature_evidence":1152,"use_case_evidence":1171,"fetched_at":1181,"status":425,"content_hash":1182},"https:\u002F\u002Fflaq.ai\u002Fmodels\u002Fgoogle\u002Fveo3-1-fast-image-to-video\u002F","Google Image To Video Veo 3.1 Fast Image-to-Video API\nVeo 3.1 Fast Image-to-Video API provides rapid Google image animation with cinematic motion, cost-effective output, and scalab","Stable & Affordable Veo 3.1 Fast Image-to-Video + Audio | Stable API for Creative Production","Rapidly animate photos via Veo 3.1 Fast API with native audio. Balanced performance and stability with an industry-leading price-to-output ratio for developers. Built for free testing and stable API workflows.",[1141,1102,1103,1104,1103,1104,1103,1104,1142,1095,1096,1143,1144,1145,1146,1147,1148,1149,1150,1151],"Related Veo3 1 Fast Models","Veo 3.1 Fast Image to Video Pricing","Fast & Affordable Veo 3.1 Fast Image-to-Video API (Google's Rapid Video Animation)","Key Features of Veo 3.1 Fast Image-to-Video API","How to Use Veo 3.1 Fast Image-to-Video API for Rapid Video Animation on Flaq AI","Best Use Cases for Veo 3.1 Fast Image-to-Video API Integration","Veo 3.1 Fast Image-to-Video vs Competitors: Comparative Analysis","Explore multiple Creative Tools with Flaq AI","AI Text to Image Generator","AI Image to Image Generator","AI Text to Video Generator",[1153,1154,1155,1156,1157,1158,1159,1156,1160,1161,1162,1163,1164,1165,1166,1167,1144,1168,1169,1170],"Stable & Affordable Veo 3.1 Fast Image-to-Video + Audio | Stable API for Creative Production Image API Video API LLM API AI Tools Free Tools Billing Resources menu Image API Image API Video API Video API MiniMax H3 MiniMax H3 Seedance 2.0 Seedance 2.0 Happy Horse 1.1 Happy Horse 1.1 Veo 3.1 Veo 3.1 Kling 3.0 Kling 3.0 Wan 2.7 Wan 2.7 Wan 2.6 Wan 2.6 Seedance 1.5 Pro Seedance 1.5 Pro Grok Imagine Video 1.5 Grok Imagine Video 1.5 Grok Imagine Video Grok Imagine Video Kling Video O3 Kling Video O3 Vidu Q3 Vidu Q3 Pixverse C1 Pixverse C1 Pixverse V6 Pixverse V6 Wan 2.7 Lora Wan 2.7 Lora Kling 3.0 Turbo Kling 3.0 Turbo Seedance 2.5 Seedance 2.5 Happy Horse 1.0 Happy Horse 1.0 Seedance 2.0 Mini Seedance 2.0 Mini Video Upscaler Video Upscaler FLUX 3 FLUX 3 Wan 3.0 Wan 3.0 LLM API LLM API AI Tools AI Tools Free Tools Free Tools Billing Billing Resources Resources Google \u002F veo3.1-fast-image-to-video veo3.1-fast-image-to-video Rapidly animate photos via Veo 3.1 Fast API with native audio. Balanced performance and stability with an industry-leading price-to-output ratio for developers. Built for free testing and stable API workflows.","veo3.1-fast-text-to-video text-to-video veo3.1-fast-image-to-video image-to-video Documentation Playground API ALL History Documentation API Examples","const response = await fetch ( 'https:\u002F\u002Fapi.flaq.ai\u002Fapi\u002Fv1\u002Fvideo\u002Ftask' , {","'Authorization' : 'Bearer YOUR_API_KEY'","const res = await fetch ( `https:\u002F\u002Fapi.flaq.ai\u002Fapi\u002Fv1\u002Fvideo\u002F ${taskId} ` , {","headers : { 'Authorization' : 'Bearer YOUR_API_KEY' }","'https:\u002F\u002Fapi.flaq.ai\u002Fapi\u002Fv1\u002Fvideo\u002Ftask' ,","poll_url = f\"https:\u002F\u002Fapi.flaq.ai\u002Fapi\u002Fv1\u002Fvideo\u002F {task_id} \"","poll_result = requests.get(poll_url, headers={ 'Authorization' : 'Bearer YOUR_API_KEY' }).json()","curl -X POST https:\u002F\u002Fapi.flaq.ai\u002Fapi\u002Fv1\u002Fvideo\u002Ftask \\","-H \"Authorization: Bearer YOUR_API_KEY\" \\","curl -X GET \"https:\u002F\u002Fapi.flaq.ai\u002Fapi\u002Fv1\u002Fvideo\u002F{task_id}\" \\","-H \"Authorization: Bearer YOUR_API_KEY\"","Fast & Affordable Veo 3.1 Fast Image-to-Video API (Google's Rapid Video Animation) Key Features of Veo 3.1 Fast Image-to-Video API How to Use Veo 3.1 Fast Image-to-Video API for Rapid Video Animation on Flaq AI Best Use Cases for Veo 3.1 Fast Image-to-Video API Integration Veo 3.1 Fast Image-to-Video vs Competitors: Comparative Analysis Fast & Affordable Veo 3.1 Fast Image-to-Video API (Google's Rapid Video Animation)","Google Veo 3.1 Fast Image-to-Video API delivers cost-effective, high-speed AI video animation for developers and creative teams. This lightning-fast video animation API integration helps you transform static images into professional video clips at 3x faster speeds and significantly lower cost compared to standard Veo 3.1. Built on Google DeepMind's optimized video synthesis architecture, the Veo 3.1 Fast model provides rapid motion understanding, while the API provides stable integration for high-volume workflows on Flaq AI.","Lightning-Fast Video Animation: Transform static images into high-quality videos 3x faster than standard Veo 3.1 through optimized processing architecture. Perfect for rapid prototyping and high-volume animation workflows.","Cost-Effective API Pricing: Save up to 62.5% compared to standard Veo 3.1 while maintaining professional quality. This affordable video animation API maximizes ROI for developers.","High-Fidelity Video Output: Animate images into production-ready videos with smooth motion, realistic dynamics, and clean transitions through cost-effective Veo API integration.",[1153,1172,1166,1167,1173,1174,1146,1175,1176,1177,1178,1179,1180],"Veo 3.1 Fast Image to Video Start Frame \u002F End Frame Start Frame prompt Translate copy 0 \u002F 2000 Optimize Prompt ratio resolution Generate Loading... loading... Try the AI Video Generator now Veo 3.1 Fast Image to Video Pricing","Lightning-Fast Video Animation: Transform static images into high-quality videos 3x faster than standard Veo 3.1 through optimized processing architecture. Perfect for rapid prototyping and high-volume animation workflows. ","Rapid Iteration Support: Flash-speed animation enables quick creative exploration, A\u002FB testing, and iterative refinement—ideal for agile development and content creation workflows. ","Rapid Prototyping & Iteration: Quickly explore animation concepts, test multiple variations, and iterate on video ideas with lightning-fast generation speeds ideal for agile workflows. ","Veo 3.1 Standard delivers maximum quality with advanced features and extended capabilities, ideal for premium commercial work. Veo 3.1 Fast prioritizes speed and cost-efficiency, animating images 3x faster at roughly 62.5% lower cost while maintaining professional quality—perfect for high-volume workflows and rapid iteration.","Pika excels at stylized animations and user-friendly interface. Veo 3.1 Fast API offers programmatic access, 3x faster generation, predictable pricing, and seamless integration into production workflows—making it superior for developers requiring scalable, fast, and affordable image animation.","Explore several AI creation tools for quick image and video workflows in your browser, then scale successful ideas with Flaq AI's production-ready model APIs. Flaq AI provides a unified API layer for all models, making it easy to use and scale your workflows.","Create polished images from prompts with leading AI image models, flexible settings, and a fast browser-based workflow.","Track Gemini Omni, Veo 4 search intent, and Veo 3.1 access. Learn what Google Gemini video creators can test now on Flaq AI for practical API workflows today.","2026-08-09T01:04:04.060Z","9a333e8704299855c873946ca807472a68a04237dbd55f7cc77964f7091cd53d",{"url":1184,"kind":1087,"anchor":1185,"title":1186,"description":1187,"headings":1188,"feature_evidence":1209,"use_case_evidence":1217,"fetched_at":1223,"status":425,"content_hash":1224},"https:\u002F\u002Fflaq.ai\u002Ftext-to-image\u002F","AI Text to Image Generator\nCreate polished images from prompts with leading AI image models, flexible settings, and a fast browser-based workflow.","Free AI Text to Image Generator With Top Models | Free Trial, Better Results & Creative Workflows","Free to create AI images from text prompts with Flaq AI. Choose top image models, adjust ratio and resolution, preview credit cost, and generate polished visuals fast.",[1189,1190,1191,1192,1193,1194,1195,1196,1197,1198,1199,1200,1201,1202,1203,1204,1205,1206,1207,1208],"AI Text to Image Generator: Create Images From Prompts","Create high-quality visuals from simple text prompts with Flaq AI Text to Image Generator. Start with free credits, choose leading image models, fine-tune output settings, and generate polished images for design, marketing, storytelling, education, and creative production.","Core Features of AI Text to Image Generator","Example of Text-to-Image Creation","Example of Creative Image Variations","Example of Style and Layout Control","Text-to-Image Generator Features","Leading Multi-Model Image Generation","Prompt-Driven Visual Control","Flexible Ratios and Output Sizes","Clear Credit Cost Before Generation","Text-to-Image Use Cases","Marketing and Social Visuals","Design Drafting and Brand Exploration","Education and Presentation Images","Storytelling and Concept Art","How to Generate Images From Text","Input Your Image Prompt","Choose Model and Settings","Generate, Preview, and Refine",[1210,1211,1212,1213,1214,1215,1216],"Free AI Text to Image Generator With Top Models | Free Trial, Better Results & Creative Workflows Image API Video API LLM API AI Tools Free Tools Billing Resources menu Image API Image API Video API Video API LLM API LLM API AI Tools AI Tools AI Media Creator AI Media Creator Reference to Video Reference to Video Text to Image Text to Image Image to Image Image to Image Text to Video Text to Video Image to Video Image to Video Claude Code Guide Claude Code Guide OpenAI Codex Guide OpenAI Codex Guide Hermes Agent Guide Hermes Agent Guide Free Tools Free Tools Billing Billing Resources Resources Text to Image Model Show labels gpt-image-2 Enter Prompt Translate copy 0 \u002F 2000 Optimize Prompt Quality quality ratio resolution Generate Loading... loading... View Details PNG History No history yet AI Text to Image Generator: Create Images From Prompts","Generate professional-grade AI images with natural-language prompts, free credits, multi-model access, aspect ratio control, resolution options, and transparent credit estimates in one creation flow.","Discover more image, video, and chat models on Flaq AI, with model pages that include capabilities, pricing details, API examples, and playground access.","AI Image API","Nano Banana Pro Nano Banana 2 GPT Image 2 Seedream 5.0 Pro Seedream 5.0 Seedream 4.5 Grok Imagine Qwen Image 2.0 Wan 2.7 Image Qwen Image Lora Z Image Qwen Image 3.0 Pro AI Video API","MiniMax H3 Seedance 2.0 Happy Horse 1.1 Veo 3.1 Kling 3.0 Wan 2.7 Wan 2.6 Seedance 1.5 Pro Grok Imagine Video 1.5 Grok Imagine Video Kling Video O3 Vidu Q3 Pixverse C1 Pixverse V6 Wan 2.7 Lora Kling 3.0 Turbo Seedance 2.5 Happy Horse 1.0 Seedance 2.0 Mini Video Upscaler FLUX 3 Wan 3.0 AI LLM API","GPT 5.6 Sol GPT 5.4 GPT 5.5 Claude Sonnet 5 Claude Sonnet 4.6 Claude Opus 4.6 Claude Opus 4.7 Text Filter API Image Filter API Claude Opus 4.8 Gemini 3.5 Flash Qwen 3.7 Grok 4.5 Claude Fable 5 Kimi K3 Kimi 2.7 GLM 5.2 Qwen Character DeepSeek v4 Grok 4 GPT 5.6 Terra GPT 5.6 Luna Claude Opus 5 Qwen 3.8 Max GLM 5.3 AI Tools",[1210,1218,1211,1219,1220,1200,1221,1222],"A powerful AI image creation workflow for turning written ideas into detailed visuals, built for strong prompt understanding, flexible model selection, clean composition, and fast creative iteration.","Create square images, vertical posts, posters, thumbnails, banners, product scenes, and widescreen concepts with model-specific aspect ratio and resolution settings.","Review estimated generation cost before you start, with free credits and pricing adapted to the selected provider, model, quality, resolution, and current discount rules.","Use Flaq AI Text to Image Generator whenever you need original images from words, without preparing a reference image or building a complex design workflow.","Select a provider and model for your creative goal, then set ratio, resolution, quality, seed, or prompt optimization when the chosen model supports them.","2026-08-09T01:04:04.850Z","a601da7198d795c9cb7d00e4f71cefcac2389e5784525e37f4e9e3056b438224",{"url":1226,"kind":484,"anchor":1227,"title":1228,"description":1229,"headings":1230,"feature_evidence":1241,"use_case_evidence":1259,"fetched_at":1271,"status":425,"content_hash":1272},"https:\u002F\u002Fflaq.ai\u002Fmodels\u002Fopenai\u002Fgpt-image-2\u002F","Try Image API","Affordable OpenAI GPT Image 2 API: ChatGPT Image 2.0","High-quality image generation by OpenAI GPT Image 2 API with flexible quality controls and strong prompt adherence. Stable and affordable for production workflows. Built for free testing and stable API workflows.",[1231,1232,1094,1095,1096,1233,1234,1235,1236,1237,1102,1103,1104,1103,1104,1103,1104,1238,1239,1240],"Related Gpt Image 2 Models","GPT Image 2 Pricing","Fast & Affordable GPT Image 2 API (OpenAI's Advanced Text-to-Image Generation)","Key Features of GPT Image 2 API","How to Use GPT Image 2 API for Professional Image Generation on Flaq AI","Best Use Cases for GPT Image 2 API Integration","GPT Image 2 vs Competitors: Comparative Analysis","More Articles for GPT Image 2","Image to Image Creation Guide: How to Edit, Restyle, and Rebuild Visuals with Flaq AI","ChatGPT Image 2 Is Here: A Practical Review of OpenAI’s New Image Model and API Access",[1242,1243,1244,1245,1234,1246,1247,1248,1249,1235,1250,1251,1252,1236,1253,1254,1255,1256,1257,1258],"Affordable OpenAI GPT Image 2 API: ChatGPT Image 2.0 Image API Video API LLM API AI Tools Free Tools Billing Resources menu Image API Image API Nano Banana Pro Nano Banana Pro Nano Banana 2 Nano Banana 2 GPT Image 2 GPT Image 2 Seedream 5.0 Pro Seedream 5.0 Pro Seedream 5.0 Seedream 5.0 Seedream 4.5 Seedream 4.5 Grok Imagine Grok Imagine Qwen Image 2.0 Qwen Image 2.0 Wan 2.7 Image Wan 2.7 Image Qwen Image Lora Qwen Image Lora Z Image Z Image Qwen Image 3.0 Pro Qwen Image 3.0 Pro Video API Video API LLM API LLM API AI Tools AI Tools Free Tools Free Tools Free Nano Banana AI Free Nano Banana AI Free GPT Image 2 Free GPT Image 2 Free Text to Image Free Text to Image Free Image to Image Free Image to Image Free Image Editor Free Image Editor Billing Billing Resources Resources OpenAI \u002F gpt-image-2 gpt-image-2 High-quality image generation by OpenAI GPT Image 2 API with flexible quality controls and strong prompt adherence. Stable and affordable for production workflows. Built for free testing and stable API workflows.","gpt-image-2 text-to-image gpt-image-2-edit image-to-image gpt-image-2-client text-to-image gpt-image-2-edit-client image-to-image Documentation Playground API ALL History Documentation GPT Image 2 Enter Prompt Translate copy 0 \u002F 2000 Optimize Prompt Quality quality ratio resolution Generate Loading... loading... Try the AI Image Generator now","Fast & Affordable GPT Image 2 API (OpenAI's Advanced Text-to-Image Generation) Key Features of GPT Image 2 API How to Use GPT Image 2 API for Professional Image Generation on Flaq AI Best Use Cases for GPT Image 2 API Integration GPT Image 2 vs Competitors: Comparative Analysis Fast & Affordable GPT Image 2 API (OpenAI's Advanced Text-to-Image Generation)","GPT Image 2 API from OpenAI delivers high-quality, cost-effective AI image generation for developers and creative teams. This professional text-to-image API integration helps you turn natural-language prompts into polished visuals with strong instruction following, flexible quality controls, and reliable output consistency. Built on OpenAI's latest GPT image generation stack, GPT Image 2 combines semantic prompt understanding with production-ready API integration for scalable creative workflows on Flaq AI.","High-Fidelity Text-to-Image Generation: Produce detailed, visually coherent images from natural-language prompts with realistic lighting, strong composition, and accurate subject rendering through GPT Image 2 API integration.","Flexible Quality Controls: Choose from multiple quality modes to balance speed and output fidelity for different production goals, making GPT Image 2 suitable for everything from rapid iteration to professional visual delivery.","Clean Text Rendering in Images: Generate marketing graphics, posters, and branded assets with clearer on-image typography and more dependable placement through OpenAI image API integration.","Production-Friendly API Integration: Power high-volume creative workflows with stable text-to-image generation, predictable controls, and professional output quality through GPT Image 2 API on Flaq AI.","Input: Natural-language text prompts with scene descriptions, style references, composition guidance, lighting details, and subject instructions.","Output: High-resolution generated images delivered via secure CDN URLs through GPT Image 2 API integration.","Capabilities: Text-to-image generation, multi-object composition, on-image typography, cinematic prompt control, and strong instruction-following through OpenAI GPT Image 2 API integration.","Marketing & Advertising: Create campaign visuals, hero banners, ad creatives, and branded assets with strong prompt accuracy through professional GPT Image 2 API workflows.","E-commerce Product Visuals: Produce product concepts, lifestyle imagery, and promotional graphics quickly through cost-effective OpenAI image API integration for scalable retail workflows.","High-Volume Production Pipelines: Support applications that need repeatable, production-ready text-to-image generation with flexible quality selection and stable OpenAI API integration.","Runway Gen-4 Image emphasizes cinematic aesthetics and reference-aware creative workflows. GPT Image 2 API differentiates with strong text rendering, dependable instruction following, and clean API integration for scalable text-to-image production.","Stable Diffusion 3.5 offers open-model flexibility and customization opportunities. GPT Image 2 provides hassle-free API integration, better out-of-the-box prompt adherence, and less operational overhead for teams that want production-ready image generation immediately.","Qwen Image 2.0 is a strong value-oriented image model with multilingual strengths. GPT Image 2 API stands out with OpenAI's polished instruction following, reliable text-in-image rendering, and a streamlined professional workflow for global creative teams.",[1242,1243,1260,1244,1245,1261,1262,1236,1263,1264,1265,1256,1258,1266,1267,1268,1269,1270],"Parameters Price Original Price Discount Resolution: 4k Quality: high $0.5280 per image $0.6600 per image 80% Resolution: 4k Quality: low $0.0240 per image $0.0300 per image 80% Resolution: 4k Quality: medium $0.1440 per image $0.1800 per image 80% Resolution: 2k Quality: high $0.3520 per image $0.4400 per image 80% Resolution: 2k Quality: low $0.0160 per image $0.0200 per image 80% Resolution: 2k Quality: medium $0.0960 per image $0.1200 per image 80% Resolution: 1k Quality: high $0.1760 per image $0.2200 per image 80% Resolution: 1k Quality: low $0.0080 per image $0.0100 per image 80% Resolution: 1k Quality: medium $0.0480 per image $0.0600 per image 80% Examples","Production-Friendly API Integration: Power high-volume creative workflows with stable text-to-image generation, predictable controls, and professional output quality through GPT Image 2 API on Flaq AI. ","Output: High-resolution generated images delivered via secure CDN URLs through GPT Image 2 API integration. ","Marketing & Advertising: Create campaign visuals, hero banners, ad creatives, and branded assets with strong prompt accuracy through professional GPT Image 2 API workflows. ","E-commerce Product Visuals: Produce product concepts, lifestyle imagery, and promotional graphics quickly through cost-effective OpenAI image API integration for scalable retail workflows. ","GPT Image 2 Client offers the same core text-to-image capabilities with a very affordable access path for teams optimizing around cost efficiency. GPT Image 2 maintains the same strong prompt adherence, text rendering, and production-ready workflow fit, making both variants suitable for professional creative applications depending on budget priorities.","Seedream 5.0 excels at high-end realism and premium visual detail. GPT Image 2 offers strong prompt control, clean typography handling, and efficient API-based integration for practical production use across a broad range of creative workflows.","resolution : '1k' ,","'resolution' : '1k' ,","\"resolution\": \"1k\",","Learn GPT Image 2 API and Edit API workflows on Flaq AI with prompt formulas, developer steps, image editing tips, and 14 copy-ready prompts for teams.","2026-08-09T01:04:05.905Z","5b126296dfe08a2901aee000fea7d8c85efd81c4fa78047c2d12609bb0ba5842",{"url":1274,"kind":484,"anchor":1275,"title":1276,"description":1277,"headings":1278,"feature_evidence":1286,"use_case_evidence":1294,"fetched_at":1309,"status":425,"content_hash":1310},"https:\u002F\u002Fflaq.ai\u002Fmodels\u002Fbytedance\u002Fseedance-2-0-image-to-video\u002F","Try Video API","Affordable Seedance 2.0 API for Image-to-Video, Realistic Humans Supported","Animate images into high-quality videos via Seedance 2.0 Image-to-Video API with realistic humans supported, built-in sound, visual fidelity, and scalable output. Built for free testing and stable API workflows.",[1279,1102,1103,1104,1103,1104,1103,1104,1280,1095,1096,1281,1282,1283,1284,1285,1148,1149,1150,1151],"Related Seedance 2.0 Models","Seedance 2.0 Image to Video Pricing","Professional Seedance V2.0 Image-to-Video API (ByteDance's Multimodal Video Animation)","Key Features of Seedance V2.0 Image-to-Video API","How to Use Seedance V2.0 Image-to-Video API for Professional Video Animation on Flaq AI","Best Use Cases for Seedance V2.0 Image-to-Video API Integration","Seedance V2.0 Image-to-Video vs Competitors: Comparative Analysis",[1287,1288,1155,1156,1157,1158,1159,1156,1160,1161,1162,1163,1164,1165,1289,1290,1282,1291,1292,1293],"Affordable Seedance 2.0 API for Image-to-Video, Realistic Humans Supported Image API Video API LLM API AI Tools Free Tools Billing Resources menu Image API Image API Video API Video API MiniMax H3 MiniMax H3 Seedance 2.0 Seedance 2.0 Happy Horse 1.1 Happy Horse 1.1 Veo 3.1 Veo 3.1 Kling 3.0 Kling 3.0 Wan 2.7 Wan 2.7 Wan 2.6 Wan 2.6 Seedance 1.5 Pro Seedance 1.5 Pro Grok Imagine Video 1.5 Grok Imagine Video 1.5 Grok Imagine Video Grok Imagine Video Kling Video O3 Kling Video O3 Vidu Q3 Vidu Q3 Pixverse C1 Pixverse C1 Pixverse V6 Pixverse V6 Wan 2.7 Lora Wan 2.7 Lora Kling 3.0 Turbo Kling 3.0 Turbo Seedance 2.5 Seedance 2.5 Happy Horse 1.0 Happy Horse 1.0 Seedance 2.0 Mini Seedance 2.0 Mini Video Upscaler Video Upscaler FLUX 3 FLUX 3 Wan 3.0 Wan 3.0 LLM API LLM API AI Tools AI Tools Free Tools Free Tools Billing Billing Resources Resources ByteDance \u002F seedance-v2.0-image-to-video seedance-v2.0-image-to-video Animate images into high-quality videos via Seedance 2.0 Image-to-Video API with realistic humans supported, built-in sound, visual fidelity, and scalable output. Built for free testing and stable API workflows.","seedance-v2.0-text-to-video text-to-video seedance-v2.0-image-to-video image-to-video seedance-v2.0-reference-to-video reference-to-video Documentation Playground API ALL History Documentation API Examples","Professional Seedance V2.0 Image-to-Video API (ByteDance's Multimodal Video Animation) Key Features of Seedance V2.0 Image-to-Video API How to Use Seedance V2.0 Image-to-Video API for Professional Video Animation on Flaq AI Best Use Cases for Seedance V2.0 Image-to-Video API Integration Seedance V2.0 Image-to-Video vs Competitors: Comparative Analysis Professional Seedance V2.0 Image-to-Video API (ByteDance's Multimodal Video Animation)","ByteDance Seedance V2.0 Image-to-Video API delivers production-grade AI video animation for developers and creative teams. This advanced image-to-video API integration enables you to transform static images into high-quality animated video clips at 480p and 720p resolutions across 4–15 second durations. Built on ByteDance's Dual-Branch Diffusion Transformer architecture, the Seedance V2.0 model combines context-aware motion synthesis with optional camera control, while the API provides stable integration for professional animation workflows on Flaq AI.","Context-Aware Motion Synthesis: The Seedance V2.0 model analyzes input images to generate natural, physics-consistent motion that preserves visual style, object relationships, and scene coherence—delivering professional animation quality through the API.","Flexible Resolution Tiers: Choose between 480p for cost-effective high-volume animation and 720p for enhanced visual quality through the same API integration.","Optional Fixed Camera Mode: Lock the camera position for stable, static-shot animations through the Seedance V2.0 Image-to-Video API—ideal for product showcases, portrait animations, and controlled scene generation.",[1287,1295,1296,1297,1298,1299,1289,1290,1300,1301,1284,1302,1303,1178,1179,1304,1305,1306,1307,1308],"resolution : '720p' ,","'resolution' : '720p' ,","\"resolution\": \"720p\",","Seedance 2.0 Image to Video Start Frame \u002F End Frame Start Frame prompt Translate copy 0 \u002F 2000 Optimize Prompt Enable Sound s ratio resolution Generate Loading... loading... Try the AI Video Generator now Seedance 2.0 Image to Video Pricing","Parameters Price Original Price Discount Resolution: 1080p $0.4800 per second $0.6000 per second 80% Resolution: 720p $0.1920 per second $0.2400 per second 80% Resolution: 4k $0.9600 per second $1.2000 per second 80% Resolution: 480p $0.0960 per second $0.1200 per second 80% Related Models","Flexible Resolution Tiers: Choose between 480p for cost-effective high-volume animation and 720p for enhanced visual quality through the same API integration. ","Resolution: 2 quality tiers — 480p (cost-effective) and 720p (premium quality) ","High-Volume Content Pipelines: Power automated image animation workflows with flexible duration controls and broad aspect ratio support for scalable production at low cost. ","Seedance 1.5 Pro offers ultra-affordable pricing for maximum volume. Seedance V2.0 Image-to-Video API advances with the Dual-Branch Diffusion Transformer architecture, extended 15-second duration support, higher 720p resolution tier, and optional fixed camera control—delivering significantly improved motion quality and visual fidelity for professional animation workflows.","Compare Higgsfield MCP and CLI with Flaq AI’s Seedance 2.0 REST endpoints, pricing, task polling, model variants, testing, and practical production use cases.","Compare WaveSpeed and Flaq AI for Seedance 2.0 API workflows, endpoint choices, prompt testing, latency, cost, fallback routing, and scalable video builds.","Explore the Seedance 2.5 release outlook, likely API upgrades, Seedance 2.0 comparison points, prompt tests, and Flaq AI workflow planning for video builders.","Seedance 2.5 release Seedance 2.5 API Seedance 2.5 vs Seedance 2.0 Flaq AI, AI video API Seedance 2.0 Fast API Read More 2026-04-16 Seedance 2.0 API Guide: How to Use Flaq AI for Faster Text-to-Video Workflows","A practical guide to the Seedance 2.0 API on Flaq AI, with features, workflow tips, pricing context, and ideas for faster text-to-video creation.","2026-08-09T01:04:06.963Z","f7f4fdcc91e23fe439225f86970914b278a3e2da79a2a7df2c83ab55e7a227f3",{"url":1312,"kind":484,"anchor":1313,"title":1314,"description":1315,"headings":1316,"feature_evidence":1325,"use_case_evidence":1338,"fetched_at":1348,"status":425,"content_hash":1349},"https:\u002F\u002Fflaq.ai\u002Fmodels\u002Falibaba\u002Fqwen-image-3-0-pro\u002F","Qwen Image 3.0 Pro API\nText To Image\nFrom $0.0350 \u002F Image","Free Qwen Image 3.0 Pro API for Text-to-Image AI Art Model","Free to try the Qwen Image 3.0 Pro API, an AI model for knowledge-rich image generation, dense layouts, fine multilingual text, realistic detail, and creative production.",[1317,1318,1094,1095,1096,1319,1320,1321,1322,1323,1102,1103,1104,1103,1104,1103,1104,1324],"Related Qwen Image 3.0 Pro Models","Qwen Image 3.0 Pro Pricing","Professional Qwen Image 3.0 Pro API (Alibaba's Advanced Image Generation)","Key Features of Qwen Image 3.0 Pro API","How to Use Qwen Image 3.0 Pro API for Professional Image Generation on Flaq AI","Best Use Cases for Qwen Image 3.0 Pro API Integration","Qwen Image 3.0 Pro vs. Competitors: Comparative Analysis","More Articles for Qwen Image 3.0 Pro",[1326,1327,1328,1329,1320,1330,1321,1331,1322,1332,1102,1333,1156,1334,1158,1335,1156,1336,1161,1337],"Free Qwen Image 3.0 Pro API for Text-to-Image AI Art Model Image API Video API LLM API AI Tools Free Tools Billing Resources menu Image API Image API Nano Banana Pro Nano Banana Pro Nano Banana 2 Nano Banana 2 GPT Image 2 GPT Image 2 Seedream 5.0 Pro Seedream 5.0 Pro Seedream 5.0 Seedream 5.0 Seedream 4.5 Seedream 4.5 Grok Imagine Grok Imagine Qwen Image 2.0 Qwen Image 2.0 Wan 2.7 Image Wan 2.7 Image Qwen Image Lora Qwen Image Lora Z Image Z Image Qwen Image 3.0 Pro Qwen Image 3.0 Pro Video API Video API LLM API LLM API AI Tools AI Tools Free Tools Free Tools Billing Billing Resources Resources Alibaba \u002F qwen-image-3.0-pro qwen-image-3.0-pro Free to try the Qwen Image 3.0 Pro API, an AI model for knowledge-rich image generation, dense layouts, fine multilingual text, realistic detail, and creative production.","qwen-image-3.0-pro text-to-image qwen-image-3.0-pro-edit image-to-image Documentation Playground API ALL History Documentation Qwen Image 3.0 Pro Enter Prompt Translate copy 0 \u002F 2000 Optimize Prompt Seed ratio resolution Generate Loading... loading... Try the AI Image Generator now","Professional Qwen Image 3.0 Pro API (Alibaba's Advanced Image Generation) Key Features of Qwen Image 3.0 Pro API How to Use Qwen Image 3.0 Pro API for Professional Image Generation on Flaq AI Best Use Cases for Qwen Image 3.0 Pro API Integration Qwen Image 3.0 Pro vs. Competitors: Comparative Analysis Professional Qwen Image 3.0 Pro API (Alibaba's Advanced Image Generation)","Qwen Image 3.0 Pro API turns natural-language prompts into polished images for creative, commercial, editorial, and product workflows. The model provides a text-to-image generation path with flexible aspect-ratio controls and seed support, while Flaq AI provides an application-ready interface for repeatable image production.","Production-Ready Image Workflow: Generate polished assets through an API-oriented process that fits review, storage, and publishing pipelines.","Input: A natural-language prompt describing the subject, composition, style, lighting, and required visual details.","Qwen Image 3.0 Pro vs. FLUX: FLUX models are popular for high-quality image generation and creative experimentation. Qwen Image 3.0 Pro provides another API-friendly choice for structured prompt-based image creation.","const response = await fetch ( 'https:\u002F\u002Fapi.flaq.ai\u002Fapi\u002Fv1\u002Fimage\u002Ftask' , {","const res = await fetch ( `https:\u002F\u002Fapi.flaq.ai\u002Fapi\u002Fv1\u002Fimage\u002F ${taskId} ` , {","'https:\u002F\u002Fapi.flaq.ai\u002Fapi\u002Fv1\u002Fimage\u002Ftask' ,","poll_url = f\"https:\u002F\u002Fapi.flaq.ai\u002Fapi\u002Fv1\u002Fimage\u002F {task_id} \"","curl -X POST https:\u002F\u002Fapi.flaq.ai\u002Fapi\u002Fv1\u002Fimage\u002Ftask \\",[1327,1339,1328,1329,1340,1341,1330,1342,1322,1343,1344,1345,1346,1347],"Parameters Price Original Price Discount Resolution: 2k $0.0700 per image - Standard Resolution: 1k $0.0350 per image - Standard Examples","Flexible Creative Styles: Explore photorealistic, editorial, illustrative, product, and cinematic directions from the same image generation workflow.","Seed-Based Reproducibility: Use seed control when a workflow needs repeatable experimentation or controlled visual variations.","Output: Generated images delivered through the Flaq AI image workflow for review, download, or downstream design work.","Qwen Image 3.0 Pro vs. Nano Banana Pro: Nano Banana Pro is strong in instruction-following and multimodal image workflows. Qwen Image 3.0 Pro provides an Alibaba-oriented text-to-image path with flexible composition and style exploration.","Qwen Image 3.0 Pro vs. Seedream: Seedream supports fast commercial image generation and editing. Qwen Image 3.0 Pro offers an alternative workflow for prompt-led visual production with seed and ratio controls.","resolution : '2k' ,","'resolution' : '2k' ,","\"resolution\": \"2k\",","2026-08-09T01:04:08.159Z","f0dad08cf0c09fef9928c7ebfc0244eda4f67911aeb469e68fcded6cc7d73b31",{"url":1351,"kind":484,"title":1352,"description":1353,"feature_evidence":1354,"use_case_evidence":1369,"fetched_at":1372,"status":425,"content_hash":1373},"https:\u002F\u002Fflaq.ai\u002Fclaude-code-guide\u002F","Practical Guide for Claude Code CLI with Flaq AI API & Skills","Practical guide for using Claude Code CLI with Flaq AI's Claude-compatible API services for faster, lower-cost API workflows. Quickly configure models and explore AI agent skills.",[1355,1356,1357,1358,1359,1360,1361,1362,1363,1364,1365,1366,1367,1368,1214,1215,1216],"Practical Guide for Claude Code CLI with Flaq AI API & Skills Image API Video API LLM API AI Tools Free Tools Billing Resources menu Image API Image API Video API Video API LLM API LLM API AI Tools AI Tools AI Media Creator AI Media Creator Reference to Video Reference to Video Text to Image Text to Image Image to Image Image to Image Text to Video Text to Video Image to Video Image to Video Claude Code Guide Claude Code Guide OpenAI Codex Guide OpenAI Codex Guide Hermes Agent Guide Hermes Agent Guide Free Tools Free Tools Billing Billing Resources Resources Flaq AI Agent Guide","Practical Guide for Using Claude Code CLI with Flaq AI's API Services and Open-Source AI Agent Skills","Use Flaq AI's Claude Code-compatible endpoint to quickly configure Claude Code CLI with Flaq and access our model services from the terminal. After configuring your API key, follow the examples in this guide to choose a currently supported model and verify the setup. You can also explore reusable open-source Claude Code skills to extend AI agent capabilities and support additional workflows.","Get API Key View Configuration macOS \u002F Linux \u002F WSL Agent guides","Claude Code \u002Fcc-api OpenAI Codex \u002Fcodex-api\u002Fv1 Hermes Agent \u002Fapi\u002Fv1 Install","Choose a model currently supported by the Flaq Claude Code endpoint, add your Flaq API key as an Anthropic auth token, and point Claude Code CLI to the Flaq API base URL. Open-source Claude Code skills cover reusable workflows for coding, deployment, testing, documents, creative work, and other tasks.","Copy export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-sonnet-4-6\" For a one-off run, you can skip ANTHROPIC_MODEL and use claude --model claude-sonnet-4-6 instead.","Copy export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-6\" For a one-off run, you can skip ANTHROPIC_MODEL and use claude --model claude-opus-4-6 instead.","Copy export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-7\" For a one-off run, you can skip ANTHROPIC_MODEL and use claude --model claude-opus-4-7 instead.","Copy $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-sonnet-4-6\" For a one-off run, you can skip ANTHROPIC_MODEL and use claude --model claude-sonnet-4-6 instead.","Copy $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-6\" For a one-off run, you can skip ANTHROPIC_MODEL and use claude --model claude-opus-4-6 instead.","Copy $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-7\" For a one-off run, you can skip ANTHROPIC_MODEL and use claude --model claude-opus-4-7 instead.","Should I use ANTHROPIC_AUTH_TOKEN or ANTHROPIC_API_KEY?","Using this guide's setup may significantly increase token usage. Install Install Claude Code CLI Configure Connect Claude Code CLI to Flaq AI Verify Verify Flaq in Claude Code CLI Skills Extend Claude Code with Open-Source AI Agent Skills Troubleshooting Claude Code FAQ AI Image API",[1357,1360,1370,1371],"claude-sonnet-4-6 claude-opus-4-6 Recommended claude-opus-4-7 Balanced Claude model for everyday coding workflows, refactors, and terminal assistance.","Claude Code skills are reusable packages of instructions, workflows, scripts, and reference files that extend AI agent capabilities across recurring tasks. Explore skills for coding, deployment, testing, documents, creative work, and other workflows without repeating the same instructions in every session. Review each skill's source, required tools, and permissions before using it.","2026-08-09T01:04:09.406Z","ee4acca9f914afb8e797a2c736589ce89783e8076169b59e0fa583befbd7bc1b",{"url":1375,"kind":484,"title":1376,"description":1377,"feature_evidence":1378,"use_case_evidence":1395,"fetched_at":1396,"status":425,"content_hash":1397},"https:\u002F\u002Fflaq.ai\u002Fja\u002Fclaude-code-guide\u002F","Claude Code CLI向けFlaq AI API接続・オープンソーススキルガイド｜低コストで安定した端末ワークフロー実践設定","Claude Code CLIでFlaq AI APIを使う実践ガイド。Claude互換エンドポイント、API key、モデル選択、ターミナル検証、接続確認を押さえ、低コストで安定したAPIワークフローを素早く構築し、複数環境での利用、コスト確認、本番前のモデル検証と認証確認に役立ち、接続確認や運用改善にも有用です。さらに、オープンソースのClaude Codeスキルを探して、コーディング、デプロイ、テスト、ドキュメント、クリエイティブ作業のワークフローを拡張できます。",[1379,1380,1381,1382,1383,1384,1385,1386,1387,1388,1389,1390,1391,1392,1393,1394],"Claude Code CLI向けFlaq AI API接続・オープンソーススキルガイド｜低コストで安定した端末ワークフロー実践設定 画像API 動画API 大規模言語モデルAPI AIツール 無料ツール 請求 リソース menu 画像API 画像API 動画API 動画API 大規模言語モデルAPI 大規模言語モデルAPI AIツール AIツール AIメディアクリエイター AIメディアクリエイター 参照素材から動画生成 参照素材から動画生成 テキストから画像 テキストから画像 画像から画像 画像から画像 テキストから動画 テキストから動画 画像から動画 画像から動画 Claude Code ガイド Claude Code ガイド OpenAI Codex ガイド OpenAI Codex ガイド Hermes Agent ガイド Hermes Agent ガイド 無料ツール 無料ツール 請求 請求 リソース リソース Flaq AI エージェントガイド","Flaq AI の API サービスとオープンソース AI エージェントスキルで Claude Code CLI を使う実践ガイド","Flaq AI の Claude Code 互換エンドポイントを使うと、Claude Code CLI を Flaq 向けにすばやく設定し、ターミナルからモデルサービスにアクセスできます。API キーを設定したら、このガイドの例に沿って現在サポートされているモデルを選び、セットアップを検証してください。再利用可能な オープンソース Claude Code スキル を探して、AI エージェントの機能を拡張し、さらに多くのワークフローに対応することもできます。","API Key を取得 設定を表示 macOS \u002F Linux \u002F WSL エージェントガイド","Claude Code \u002Fcc-api OpenAI Codex \u002Fcodex-api\u002Fv1 Hermes Agent \u002Fapi\u002Fv1 インストール","Flaq Claude Code エンドポイントで現在サポートされているモデルを選び、Flaq API キーを Anthropic 認証トークンとして追加し、Claude Code CLI の向き先を Flaq API ベース URL に設定します。 オープンソース Claude Code スキル には、コーディング、デプロイ、テスト、ドキュメント、クリエイティブ作業などのタスクに対応する再利用可能なワークフローがあります。","コピー export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-sonnet-4-6\" 1 回だけ実行する場合は、 ANTHROPIC_MODEL を省略し、代わりに claude --model claude-sonnet-4-6 を使えます。","コピー export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-6\" 1 回だけ実行する場合は、 ANTHROPIC_MODEL を省略し、代わりに claude --model claude-opus-4-6 を使えます。","コピー export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-7\" 1 回だけ実行する場合は、 ANTHROPIC_MODEL を省略し、代わりに claude --model claude-opus-4-7 を使えます。","コピー $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-sonnet-4-6\" 1 回だけ実行する場合は、 ANTHROPIC_MODEL を省略し、代わりに claude --model claude-sonnet-4-6 を使えます。","コピー $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-6\" 1 回だけ実行する場合は、 ANTHROPIC_MODEL を省略し、代わりに claude --model claude-opus-4-6 を使えます。","コピー $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-7\" 1 回だけ実行する場合は、 ANTHROPIC_MODEL を省略し、代わりに claude --model claude-opus-4-7 を使えます。","ANTHROPIC_AUTH_TOKEN と ANTHROPIC_API_KEY のどちらを使うべきですか？","このガイドの設定で使用すると、token 使用量が大幅に増える可能性があります。 インストール Claude Code CLI をインストールする 設定 Claude Code CLI を Flaq AI に接続する 検証 Claude Code CLI で Flaq を検証する スキル オープンソース AI エージェントスキルで Claude Code を拡張する トラブルシューティング Claude Code FAQ AI画像API","ナノバナナプロ Nano Banana 2 GPT Image 2 Seedream 5.0 Pro Seedream 5.0 Seedream 4.5 Grok Imagine Qwen Image 2.0 Wan 2.7 Image Qwen Image Lora Z Image Qwen Image 3.0 Pro AI動画API","MiniMax H3 Seedance 2.0 Happy Horse 1.1 Veo 3.1 Kling 3.0 Wan 2.7 Wan 2.6 Seedance 1.5 Pro Grok Imagine Video 1.5 Grok Imagine Video Kling Video O3 Vidu Q3 Pixverse C1 Pixverse V6 Wan 2.7 Lora Kling 3.0 Turbo Seedance 2.5 Happy Horse 1.0 Seedance 2.0 Mini Video Upscaler FLUX 3 Wan 3.0 AI大規模言語モデルAPI",[],"2026-08-09T01:04:10.015Z","d7321d92efce8a8f07ee54812d0ca8a697371032b445fa671a3ac836b543cd11",{"url":1399,"kind":484,"title":1400,"description":1401,"feature_evidence":1402,"use_case_evidence":1419,"fetched_at":1420,"status":425,"content_hash":1421},"https:\u002F\u002Fflaq.ai\u002Fid\u002Fclaude-code-guide\u002F","Panduan praktis menggunakan Claude Code CLI dengan layanan API Flaq AI dan Claude Code Skills open-source","Panduan praktis menggunakan Claude Code CLI dengan layanan API Flaq AI untuk alur kerja API yang lebih cepat dan lebih hemat biaya. Gunakan endpoint Flaq AI yang kompatibel dengan Claude untuk mengonfigurasi Claude Code CLI dengan Flaq secara cepat dan mengakses layanan model kami dari terminal. Jelajahi Claude Code Skills open-source untuk memperluas alur kerja coding, deployment, pengujian, dokumentasi, dan pekerjaan kreatif.",[1403,1404,1405,1406,1407,1408,1409,1410,1411,1412,1413,1414,1415,1416,1417,1418],"Panduan praktis menggunakan Claude Code CLI dengan layanan API Flaq AI dan Claude Code Skills open-source API Gambar API Video LLM API Alat AI Alat Gratis Penagihan Sumber Daya menu API Gambar API Gambar API Video API Video LLM API LLM API Alat AI Alat AI Pembuat Media AI Pembuat Media AI Referensi ke Video Referensi ke Video Teks ke Gambar Teks ke Gambar Gambar ke Gambar Gambar ke Gambar Teks ke Video Teks ke Video Gambar ke Video Gambar ke Video Panduan Claude Code Panduan Claude Code Panduan OpenAI Codex Panduan OpenAI Codex Panduan Hermes Agent Panduan Hermes Agent Alat Gratis Alat Gratis Penagihan Penagihan Sumber Daya Sumber Daya Panduan agen Flaq AI","Panduan Praktis Menggunakan Claude Code CLI dengan Layanan API Flaq AI dan Skill Agen AI Open-Source","Gunakan endpoint Flaq AI yang kompatibel dengan Claude Code untuk mengonfigurasi Claude Code CLI dengan Flaq secara cepat dan mengakses layanan model kami dari terminal. Setelah mengonfigurasi API key, ikuti contoh dalam panduan ini untuk memilih model yang saat ini didukung dan memverifikasi setup. Anda juga dapat menjelajahi skill Claude Code open-source yang dapat digunakan kembali untuk memperluas kemampuan agen AI dan mendukung alur kerja tambahan.","Dapatkan API Key Lihat konfigurasi macOS \u002F Linux \u002F WSL Panduan agen","Claude Code \u002Fcc-api OpenAI Codex \u002Fcodex-api\u002Fv1 Hermes Agent \u002Fapi\u002Fv1 Instal","Pilih model yang saat ini didukung oleh endpoint Flaq Claude Code, tambahkan API key Flaq Anda sebagai token auth Anthropic, dan arahkan Claude Code CLI ke URL dasar API Flaq. Skill Claude Code open-source mencakup alur kerja yang dapat digunakan kembali untuk coding, deployment, testing, dokumen, pekerjaan kreatif, dan tugas lainnya.","Salin export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-sonnet-4-6\" Untuk satu kali run, Anda dapat melewati ANTHROPIC_MODEL dan menggunakan claude --model claude-sonnet-4-6 sebagai gantinya.","Salin export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-6\" Untuk satu kali run, Anda dapat melewati ANTHROPIC_MODEL dan menggunakan claude --model claude-opus-4-6 sebagai gantinya.","Salin export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-7\" Untuk satu kali run, Anda dapat melewati ANTHROPIC_MODEL dan menggunakan claude --model claude-opus-4-7 sebagai gantinya.","Salin $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-sonnet-4-6\" Untuk satu kali run, Anda dapat melewati ANTHROPIC_MODEL dan menggunakan claude --model claude-sonnet-4-6 sebagai gantinya.","Salin $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-6\" Untuk satu kali run, Anda dapat melewati ANTHROPIC_MODEL dan menggunakan claude --model claude-opus-4-6 sebagai gantinya.","Salin $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-7\" Untuk satu kali run, Anda dapat melewati ANTHROPIC_MODEL dan menggunakan claude --model claude-opus-4-7 sebagai gantinya.","Haruskah saya menggunakan ANTHROPIC_AUTH_TOKEN atau ANTHROPIC_API_KEY?","Menggunakan setup dalam panduan ini dapat meningkatkan penggunaan token secara signifikan. Instal Instal Claude Code CLI Konfigurasi Hubungkan Claude Code CLI ke Flaq AI Verifikasi Verifikasi Flaq di Claude Code CLI Keahlian Perluas Claude Code dengan Skill Agen AI Open-Source Pemecahan masalah Pertanyaan Umum Claude Code API Gambar AI","Nano Banana Pro Nano Banana 2 GPT Image 2 Seedream 5.0 Pro Seedream 5.0 Seedream 4.5 Grok Imagine Qwen Image 2.0 Wan 2.7 Image Qwen Image Lora Z Image Qwen Image 3.0 Pro API Video AI","MiniMax H3 Seedance 2.0 Happy Horse 1.1 Veo 3.1 Kling 3.0 Wan 2.7 Wan 2.6 Seedance 1.5 Pro Grok Imagine Video 1.5 Grok Imagine Video Kling Video O3 Vidu Q3 Pixverse C1 Pixverse V6 Wan 2.7 Lora Kling 3.0 Turbo Seedance 2.5 Happy Horse 1.0 Seedance 2.0 Mini Video Upscaler FLUX 3 Wan 3.0 API AI Model Bahasa Besar",[],"2026-08-09T01:04:10.609Z","f58ffcaf639c1a1c6de7d3b5af3d2dfb5ef1c8f185490f3fe2bbaa7c1082498f",{"url":1423,"kind":484,"title":1424,"description":1425,"feature_evidence":1426,"use_case_evidence":1442,"fetched_at":1445,"status":425,"content_hash":1446},"https:\u002F\u002Fflaq.ai\u002Fit\u002Fclaude-code-guide\u002F","Guida pratica all’uso di Claude Code CLI con i servizi API di Flaq AI e le skill open source di Claude Code","Guida pratica all’uso di Claude Code CLI con i servizi API di Flaq AI per workflow API più rapidi e meno costosi. Usa l’endpoint compatibile con Claude di Flaq AI per configurare rapidamente Claude Code CLI con Flaq e accedere ai nostri servizi di modelli dal terminale. Esplora inoltre le skill open source di Claude Code per ampliare i workflow di programmazione, distribuzione, test, documentazione e creatività.",[1427,1428,1429,1430,1431,1432,1433,1434,1435,1436,1433,1437,1438,1439,1440,1441,1441,1440,1441,1441],"Guida pratica all’uso di Claude Code CLI con i servizi API di Flaq AI e le skill open source di Claude Code API Immagini API Video LLM API Strumenti AI Strumenti Gratuiti Fatturazione Risorse menu API Immagini API Immagini API Video API Video LLM API LLM API Strumenti AI Strumenti AI Creatore multimediale IA Creatore multimediale IA Da riferimento a video Da riferimento a video Da testo a immagine Da testo a immagine Da immagine a immagine Da immagine a immagine Da testo a video Da testo a video Da immagine a video Da immagine a video Guida Claude Code Guida Claude Code Guida OpenAI Codex Guida OpenAI Codex Guida Hermes Agent Guida Hermes Agent Strumenti Gratuiti Strumenti Gratuiti Fatturazione Fatturazione Risorse Risorse Guida agli agenti Flaq AI","Guida pratica all'uso di Claude Code CLI con i servizi API di Flaq AI e skill open source per agenti AI","Usa l'endpoint compatibile con Claude Code di Flaq AI per configurare rapidamente Claude Code CLI con Flaq e accedere ai nostri servizi di modelli dal terminale. Dopo aver configurato la tua chiave API, segui gli esempi in questa guida per scegliere un modello attualmente supportato e verificare la configurazione. Puoi anche esplorare skill open source per Claude Code riutilizzabili per estendere le capacità degli agenti AI e supportare ulteriori workflow.","Ottieni API Key Visualizza configurazione macOS \u002F Linux \u002F WSL Guide per agenti","Claude Code \u002Fcc-api OpenAI Codex \u002Fcodex-api\u002Fv1 Hermes Agent \u002Fapi\u002Fv1 Installa","Scegli un modello attualmente supportato dall'endpoint Claude Code di Flaq, aggiungi la tua chiave API Flaq come token di autenticazione Anthropic e punta Claude Code CLI all'URL base dell'API Flaq. Le skill open source per Claude Code offrono workflow riutilizzabili per sviluppo, deployment, test, documenti, lavoro creativo e altre attività.","Connessione rapida Impostazioni globali Usa variabili d'ambiente temporanee nella sessione di terminale corrente. È il modo più rapido per provare Flaq. Non modifica ~\u002F.claude\u002Fsettings.json né il tuo profilo shell.","Copia export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-sonnet-4-6\" Per un'esecuzione una tantum, puoi saltare ANTHROPIC_MODEL e usare invece claude --model claude-sonnet-4-6 .","Copia export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-6\" Per un'esecuzione una tantum, puoi saltare ANTHROPIC_MODEL e usare invece claude --model claude-opus-4-6 .","Copia export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-7\" Per un'esecuzione una tantum, puoi saltare ANTHROPIC_MODEL e usare invece claude --model claude-opus-4-7 .","Copia $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-sonnet-4-6\" Per un'esecuzione una tantum, puoi saltare ANTHROPIC_MODEL e usare invece claude --model claude-sonnet-4-6 .","Copia $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-6\" Per un'esecuzione una tantum, puoi saltare ANTHROPIC_MODEL e usare invece claude --model claude-opus-4-6 .","Copia $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-7\" Per un'esecuzione una tantum, puoi saltare ANTHROPIC_MODEL e usare invece claude --model claude-opus-4-7 .","Copia claude Comando di test rapido","Copia claude -p \"What is 2+2?\" Comando di test rapido",[1429,1432,1443,1444],"claude-sonnet-4-6 claude-opus-4-6 Recommended claude-opus-4-7 Modello Claude bilanciato per workflow di sviluppo quotidiani, refactoring e assistenza da terminale.","Le skill di Claude Code sono pacchetti riutilizzabili di istruzioni, workflow, script e file di riferimento che estendono le capacità degli agenti AI nelle attività ricorrenti. Esplora skill per sviluppo, deployment, test, documenti, lavoro creativo e altri workflow senza ripetere le stesse istruzioni in ogni sessione. Verifica il codice sorgente, gli strumenti richiesti e le autorizzazioni di ogni skill prima di usarla.","2026-08-09T01:04:11.250Z","eb66ff8075fd7ce9a891406462e2eb185b1c969858bf6a3abb9b1e31a7a01728",{"url":1448,"kind":484,"title":1449,"description":1450,"feature_evidence":1451,"use_case_evidence":1468,"fetched_at":1469,"status":425,"content_hash":1470},"https:\u002F\u002Fflaq.ai\u002Fpt\u002Fclaude-code-guide\u002F","Guia prático para usar Claude Code CLI com os serviços API da Flaq AI e as skills de código aberto do Claude Code","Guia prático para usar Claude Code CLI com os serviços API da Flaq AI em fluxos de API mais rápidos e com menor custo. Use o endpoint compatível com Claude da Flaq AI para configurar rapidamente o Claude Code CLI com a Flaq e acessar nossos serviços de modelos pelo terminal. Explore também as skills de código aberto do Claude Code para ampliar fluxos de trabalho de programação, implantação, testes, documentação e criatividade.",[1452,1453,1454,1455,1456,1457,1458,1459,1460,1461,1462,1463,1464,1465,1466,1467],"Guia prático para usar Claude Code CLI com os serviços API da Flaq AI e as skills de código aberto do Claude Code API de Imagem API de Vídeo LLM API Ferramentas de IA Ferramentas Grátis Faturamento Recursos menu API de Imagem API de Imagem API de Vídeo API de Vídeo LLM API LLM API Ferramentas de IA Ferramentas de IA Criador de mídia com IA Criador de mídia com IA Referência para vídeo Referência para vídeo Texto para Imagem Texto para Imagem Imagem para Imagem Imagem para Imagem Texto para Vídeo Texto para Vídeo Imagem para Vídeo Imagem para Vídeo Guia Claude Code Guia Claude Code Guia OpenAI Codex Guia OpenAI Codex Guia Hermes Agent Guia Hermes Agent Ferramentas Grátis Ferramentas Grátis Faturamento Faturamento Recursos Recursos Guia de agentes da Flaq AI","Guia prático para usar a Claude Code CLI com os serviços de API da Flaq AI e skills de agentes de IA de código aberto","Use o endpoint compatível com Claude Code da Flaq AI para configurar rapidamente a Claude Code CLI com a Flaq e acessar nossos serviços de modelos pelo terminal. Depois de configurar sua chave de API, siga os exemplos deste guia para escolher um modelo atualmente compatível e verificar a configuração. Você também pode explorar skills de código aberto para Claude Code reutilizáveis para ampliar os recursos dos agentes de IA e oferecer suporte a outros fluxos de trabalho.","Obter API Key Ver configuração macOS \u002F Linux \u002F WSL Guias de agentes","Claude Code \u002Fcc-api OpenAI Codex \u002Fcodex-api\u002Fv1 Hermes Agent \u002Fapi\u002Fv1 Instalar","Escolha um modelo atualmente compatível com o endpoint Claude Code da Flaq, adicione sua chave de API da Flaq como token de autenticação da Anthropic e aponte a Claude Code CLI para a URL base da API da Flaq. Os skills de código aberto para Claude Code oferecem fluxos de trabalho reutilizáveis para programação, implantação, testes, documentos, trabalho criativo e outras tarefas.","Copiar export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-sonnet-4-6\" Para uma execução única, você pode ignorar ANTHROPIC_MODEL e usar claude --model claude-sonnet-4-6 em vez disso.","Copiar export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-6\" Para uma execução única, você pode ignorar ANTHROPIC_MODEL e usar claude --model claude-opus-4-6 em vez disso.","Copiar export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-7\" Para uma execução única, você pode ignorar ANTHROPIC_MODEL e usar claude --model claude-opus-4-7 em vez disso.","Copiar $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-sonnet-4-6\" Para uma execução única, você pode ignorar ANTHROPIC_MODEL e usar claude --model claude-sonnet-4-6 em vez disso.","Copiar $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-6\" Para uma execução única, você pode ignorar ANTHROPIC_MODEL e usar claude --model claude-opus-4-6 em vez disso.","Copiar $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-7\" Para uma execução única, você pode ignorar ANTHROPIC_MODEL e usar claude --model claude-opus-4-7 em vez disso.","Devo usar ANTHROPIC_AUTH_TOKEN ou ANTHROPIC_API_KEY?","Usar esta configuração do guia pode aumentar significativamente o uso de tokens. Instalar Instalar a Claude Code CLI Configurar Conectar a Claude Code CLI à Flaq AI Verificar Verificar a Flaq na Claude Code CLI Habilidades Ampliar o Claude Code com skills de agentes de IA de código aberto Solução de problemas FAQ da Claude Code API de Imagem IA","Nano Banana Pro Nano Banana 2 GPT Image 2 Seedream 5.0 Pro Seedream 5.0 Seedream 4.5 Grok Imagine Qwen Image 2.0 Wan 2.7 Image Qwen Image Lora Z Image Qwen Image 3.0 Pro API de Vídeo IA","MiniMax H3 Seedance 2.0 Happy Horse 1.1 Veo 3.1 Kling 3.0 Wan 2.7 Wan 2.6 Seedance 1.5 Pro Grok Imagine Video 1.5 Grok Imagine Video Kling Video O3 Vidu Q3 Pixverse C1 Pixverse V6 Wan 2.7 Lora Kling 3.0 Turbo Seedance 2.5 Happy Horse 1.0 Seedance 2.0 Mini Video Upscaler FLUX 3 Wan 3.0 API de IA para grandes modelos de linguagem",[],"2026-08-09T01:04:11.898Z","d9ed58ba34bfc5986482c7a3cb1fc52846cf00310e0413db9fd792b0de9d4395",{"url":1472,"kind":484,"title":1473,"description":1474,"feature_evidence":1475,"use_case_evidence":1491,"fetched_at":1492,"status":425,"content_hash":1493},"https:\u002F\u002Fflaq.ai\u002Fes\u002Fclaude-code-guide\u002F","Guía práctica para usar Claude Code CLI con los servicios API de Flaq AI y las skills de código abierto de Claude Code","Guía práctica para usar Claude Code CLI con los servicios API de Flaq AI en flujos de API más rápidos y de menor costo. Usa el endpoint compatible con Claude de Flaq AI para configurar rápidamente Claude Code CLI con Flaq y acceder a nuestros servicios de modelos desde la terminal. Explora también las skills de código abierto de Claude Code para ampliar los flujos de trabajo de programación, despliegue, pruebas, documentación y creatividad.",[1476,1477,1478,1479,1456,1480,1481,1482,1483,1484,1485,1486,1487,1488,1489,1490],"Guía práctica para usar Claude Code CLI con los servicios API de Flaq AI y las skills de código abierto de Claude Code API de imágenes API de videos LLM API Herramientas de IA Herramientas gratuitas Facturación Recursos menu API de imágenes API de imágenes API de videos API de videos LLM API LLM API Herramientas de IA Herramientas de IA Creador multimedia con IA Creador multimedia con IA Referencia a vídeo Referencia a vídeo Texto a imagen Texto a imagen Imagen a imagen Imagen a imagen Texto a video Texto a video Imagen a video Imagen a video Guía de Claude Code Guía de Claude Code Guía de OpenAI Codex Guía de OpenAI Codex Guía de Hermes Agent Guía de Hermes Agent Herramientas gratuitas Herramientas gratuitas Facturación Facturación Recursos Recursos Guía de agentes de Flaq AI","Guía práctica para usar Claude Code CLI con los servicios API de Flaq AI y skills de agentes de IA de código abierto","Usa el endpoint compatible con Claude Code de Flaq AI para configurar rápidamente Claude Code CLI con Flaq y acceder a nuestros servicios de modelos desde la terminal. Después de configurar tu clave de API, sigue los ejemplos de esta guía para elegir un modelo compatible actualmente y verificar la configuración. También puedes explorar skills de Claude Code de código abierto reutilizables para ampliar las capacidades de los agentes de IA y admitir flujos de trabajo adicionales.","Obtener API Key Ver configuración macOS \u002F Linux \u002F WSL Guías de agentes","Elige un modelo compatible actualmente con el endpoint Claude Code de Flaq, añade tu clave de API de Flaq como token de autenticación de Anthropic y apunta Claude Code CLI a la URL base de la API de Flaq. Los skills de Claude Code de código abierto ofrecen flujos de trabajo reutilizables para programación, despliegue, pruebas, documentos, trabajo creativo y otras tareas.","Copiar export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-sonnet-4-6\" Para una ejecución puntual, puedes omitir ANTHROPIC_MODEL y usar claude --model claude-sonnet-4-6 en su lugar.","Copiar export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-6\" Para una ejecución puntual, puedes omitir ANTHROPIC_MODEL y usar claude --model claude-opus-4-6 en su lugar.","Copiar export ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" export ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" export ANTHROPIC_MODEL = \"claude-opus-4-7\" Para una ejecución puntual, puedes omitir ANTHROPIC_MODEL y usar claude --model claude-opus-4-7 en su lugar.","Copiar $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-sonnet-4-6\" Para una ejecución puntual, puedes omitir ANTHROPIC_MODEL y usar claude --model claude-sonnet-4-6 en su lugar.","Copiar $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-6\" Para una ejecución puntual, puedes omitir ANTHROPIC_MODEL y usar claude --model claude-opus-4-6 en su lugar.","Copiar $env:ANTHROPIC_AUTH_TOKEN = \"Your_Flaq_Api_Key\" $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.flaq.ai\u002Fcc-api\" $env:ANTHROPIC_MODEL = \"claude-opus-4-7\" Para una ejecución puntual, puedes omitir ANTHROPIC_MODEL y usar claude --model claude-opus-4-7 en su lugar.","¿Debo usar ANTHROPIC_AUTH_TOKEN o ANTHROPIC_API_KEY?","Usar esta configuración de la guía puede aumentar significativamente el uso de tokens. Instalar Instalar Claude Code CLI Configurar Conectar Claude Code CLI a Flaq AI Verificar Verificar Flaq en Claude Code CLI Habilidades Ampliar Claude Code con skills de agentes de IA de código abierto Solución de problemas FAQ de Claude Code API de imágenes AI","Nano Banana Pro Nano Banana 2 GPT Image 2 Seedream 5.0 Pro Seedream 5.0 Seedream 4.5 Grok Imagine Qwen Image 2.0 Wan 2.7 Image Qwen Image Lora Z Image Qwen Image 3.0 Pro API de videos AI","MiniMax H3 Seedance 2.0 Happy Horse 1.1 Veo 3.1 Kling 3.0 Wan 2.7 Wan 2.6 Seedance 1.5 Pro Grok Imagine Video 1.5 Grok Imagine Video Kling Video O3 Vidu Q3 Pixverse C1 Pixverse V6 Wan 2.7 Lora Kling 3.0 Turbo Seedance 2.5 Happy Horse 1.0 Seedance 2.0 Mini Video Upscaler FLUX 3 Wan 3.0 API de modelos de lenguaje con IA",[],"2026-08-09T01:04:12.456Z","3449b77d5f38689dc3824d1668181e70a53cbf4b70c16615039afd9262b64283",[1495,1498,1501],{"url":1496,"title":1497,"kind":613},"https:\u002F\u002Fflaq.ai\u002Fmodels\u002Fanthropic\u002Fclaude-opus-4-8-file-analysis\u002F","Anthropic File Analysis Claude Opus 4.8 File Analysis API\nClaude Opus 4.8 File Analysis API supports deep document review, report summarization, extraction, and high-accuracy file ",{"url":1499,"title":1500,"kind":613},"https:\u002F\u002Fflaq.ai\u002Fmodels\u002Fopenai\u002Fgpt-5-5-web-search\u002F","OpenAI Web Search GPT 5.5 Web Search API\nGPT 5.5 Web Search API combines advanced OpenAI reasoning with web search for research, current information, analysis, and grounded product",{"url":1502,"title":1503,"kind":613},"https:\u002F\u002Fflaq.ai\u002Fblog\u002Fdetail\u002FKimi-K3-API-Explained-What-Developers-Should-Know-Before-Flaq-AI-Access-Arrives-7d54e15e87e0\u002F","2026-07-22 Kimi K3 API Explained: What Developers Should Know Before Flaq AI Access Arrives\nExplore Kimi K3 for coding, multimodal research, and AI agents, plus official API access",[],[1107,1108,1109,1110,1125,1126,1127,1100,1128,1129,1130,1131,1132,1153,1172,1166,1167,1173,1174,1146,1175,1176,1177,1178,1179,1180,1210,1218,1211,1219,1220,1200,1221,1222],[981,1064,902,1065,1066,904,905,1067,1068,985,986,987,988,989,990,991,992,993,926,999,1000,930,1002,1003,1004,1005,1006,1007,1008,1009,940,1011,1012,1013,945,1015,948,1017,1018,1019,1020,1021,1022,1023,1024,958,960,962,1028,964,1030,967,1032,1045,1046,1047,1048,1049,1069,1051,1052,1053,1054,1055,1070,1057,1058,1059,1071,1072,1073,1074,1075,1076,1077,1078,1079,1080,1081,1082],"tak",[],"api",[1511,1512,1513,1514,1515,1516,1517,1518,1519,1520,1521,1522,1523,1524,1525,1526,1527,1528,1529,1530,1531,1532],"Zunifikowana platforma API do modeli image, video, LLM i audio","Dostęp do modeli przez jedno API","API Docs, Model Market i przewodniki integracyjne","Przeglądanie i testowanie modeli w przeglądarce","Free tools do szybkich testów obrazów","Obsługa generowania obrazów, wideo i zadań LLM","Asynchroniczne zadania API z pollingiem odpowiedzi","Wersje modelowe i różne trasy modelowe (np. text-to-video, image-to-video, reference-to-video)","Kredytowy\u002Fbillingowy model rozliczeń widoczny na stronie pricing","Wsparcie wielu języków interfejsu","Access to a unified API for text, image, and video generation","Support for multiple AI models including GPT Image 2, Seedance, Nano Banana, Kimi K3, and Claude Opus 4.7","Generation of video content from text or images with a multimodal open-source video model","Real-time video generation at 24 FPS with 768x512 resolution","Control over characters, scenes, and visual style in generated videos","Image editing via text prompts or reference images","Video generation from images using the Grok (xAI) model","Free tools for image generation, editing, and browser-based workflows","Support for 4K image editing with SeeDream V5.0 Edit API","Fast image-to-video generation with Veo 3.1 Fast API and native audio","Video upscaling via Video Upscaler Pro API","Integration with multiple LLMs for reasoning, coding, writing, and automation workflows","Free tools available for image generation, editing, and browser-based workflows.",[1535,1536],"OpenAI Codex","Hermes Agent",[720,1538,1539,1540,1541,1542,1543,1544,1545,1546,1547,1548,1549,1550,1551],"日本語","Bahasa Indonesia","Italiano","Português","Español","Deutsch","Русский","Français","简体中文","繁體中文","한국어","ไทย","Tiếng Việt","العربية",[1553,1554,1555,1556,1557],"Nie ma potwierdzonej publicznie polityki retencji danych w zebranych materiałach","Nie ma potwierdzonej informacji o self-hostingu","Nie ma potwierdzonej informacji, że wszystkie modele wspierają audio lub wszystkie funkcje na każdej trasie","Ceny i dostępność mogą różnić się zależnie od konkretnego modelu i trasy","Nie wszystkie podstrony ujawniają pełne parametry, limity i warunki komercyjne; trzeba sprawdzać live page\u002Fdocs",[1559,1560,1561,1562,1563,1564,1565,1566,1567,1568,1569,1570,1571,1572,1573,1574,1575,1576,1577,1578,1579,1580,1581,1582,1583,1584,1585,1586,1587,1588,1589,1590,1591,1592,1593,1594,1595,1596,1597,1598,1599,1600,1601,1602,1603,1604,1605,1606,1607,1608,787,1609,1610,1611,1612,1613,1614,1615,1616,1617,1618,1619,1620,1621,939,908,1622,925,1623,957,1624],"Nano Banana Pro","Nano Banana 2","GPT Image 2","Seedream 5.0 Pro","Seedream 5.0","Seedream 4.5","Grok Imagine","Qwen Image 2.0","Wan 2.7 Image","Qwen Image Lora","Z Image","Qwen Image 3.0 Pro","MiniMax H3","Seedance 2.0","Happy Horse 1.1","Veo 3.1","Kling 3.0","Wan 2.7","Wan 2.6","Seedance 1.5 Pro","Grok Imagine Video 1.5","Grok Imagine Video","Kling Video O3","Vidu Q3","Pixverse C1","Pixverse V6","Wan 2.7 Lora","Kling 3.0 Turbo","Seedance 2.5","Happy Horse 1.0","Seedance 2.0 Mini","Video Upscaler","FLUX 3","Wan 3.0","GPT 5.6 Sol","GPT 5.4","GPT 5.5","Claude Sonnet 5","Claude Sonnet 4.6","Claude Opus 4.6","Claude Opus 4.7","Text Filter API","Image Filter API","Claude Opus 4.8","Gemini 3.5 Flash","Qwen 3.7","Grok 4.5","Claude Fable 5","Kimi K3","Kimi 2.7","Qwen Character","DeepSeek v4","Grok 4","GPT 5.6 Terra","GPT 5.6 Luna","Claude Opus 5","Qwen 3.8 Max","GLM 5.3","Seedance","Nano Banana","Grok (xAI)","SeeDream V5.0","Happy Horse","Flux","Pixverse","GLM",[1626,1631,1634,1637,1640],{"plan_name":1627,"access_model":680,"price":1628,"currency":740,"billing_period":1629,"included_usage":1630,"trial_days":680,"card_required":680,"source_evidence":680},"Select Package","$10","jednorazowe doładowanie kredytów \u002F usage-based","$10 kredytów",{"plan_name":1627,"access_model":680,"price":1632,"currency":740,"billing_period":1629,"included_usage":1633,"trial_days":680,"card_required":680,"source_evidence":680},"$20","$20 kredytów",{"plan_name":1627,"access_model":680,"price":1635,"currency":740,"billing_period":1629,"included_usage":1636,"trial_days":680,"card_required":680,"source_evidence":680},"$100","$103 kredytów (+3%)",{"plan_name":1627,"access_model":680,"price":1638,"currency":740,"billing_period":1629,"included_usage":1639,"trial_days":680,"card_required":680,"source_evidence":680},"$200","$206 kredytów (+3%)",{"plan_name":1641,"access_model":1642,"price":1643,"currency":740,"billing_period":680,"included_usage":680,"trial_days":680,"card_required":680,"source_evidence":906},"Free","freemium","0.00","Strona pricing pokazuje model oparty na kredytach i rozliczeniu zależnym od użycia. Widoczna jest też informacja o darmowym doświadczeniu próbnego startu (free trial experience) oraz saldzie konta. Dla modeli na stronie głównej widać ceny od np. $0.0050\u002Fsek. dla Video Upscaler API, a wybrane modele mają stawki per obraz \u002F per sekunda \u002F per 1M tokenów.","W zebranych oficjalnych materiałach nie ma jednoznacznej informacji, czy dane użytkowników są używane do trenowania modeli. Na stronie pojawiają się jedynie odwołania do Privacy Policy i Terms & Conditions, ale bez treści potwierdzającej politykę treningu danych.","2026",[],"Flaq AI to oficjalnie opisana jako zunifikowana platforma API do modeli image, video, audio i language, budowana przez FLAQ AI PTE. LTD. w Singapurze. Strona główna i About potwierdzają katalog modeli, dokumentację API, narzędzia browserowe, grupy docelowe oraz wiele języków interfejsu. Strona pricing pokazuje rozliczenie w kredytach i doładowaniach. W zebranych źródłach nie udało się jednak potwierdzić kilku kluczowych pól compliance, takich jak data retention region, GDPR czy trening danych, więc pozostają null.","W dostępnych materiałach nie ma pełnej, oficjalnie zweryfikowanej specyfikacji bezpieczeństwa. Strona sugeruje dokumentację, warunki i politykę prywatności, ale nie dostarcza tu dowodu na region danych, retencję, czy konkretne zabezpieczenia poza ogólnym opisem platformy i API.","nie","Zunifikowana platforma API Flaq AI do modeli image, video, audio i LLM z rozliczeniem na kredyty\u002Fzużycie oraz browserowymi narzędziami kreatywnymi.",[751,1653],"iOS",[1655,1656,1657,1658,1659,1660,1661,1662,1663,1664,1665,1666,1667,1668],"Developers & Agent Builders","Startups & Product Teams","Designers & Creative Teams","Marketing Teams","E-commerce & Brands","Creators & Publishers","Businesses & Agencies","Educators & Learners","Model Reviewers & Researchers","Developers","Creative teams","Content creators","Marketing professionals","Automation engineers","Free tools available for testing image generation and editing workflows.",[1671,1672,1673,1674,1675,1676,1677,1678,1679,1680,1681,1682,1683,1684],"Generowanie i edycja obrazów","Generowanie i animacja wideo","Zadania LLM: reasoning, coding, writing, search, file analysis","Prototypowanie funkcji AI w produktach","Automatyzacja workflow content marketingu","Testowanie modeli w przeglądarce przed wdrożeniem","Tworzenie kampanii, reklam i assetów brandowych","Integracje agentów i narzędzi terminalowych","Creating engaging YouTube, Twitch, and Facebook video thumbnails","Generating realistic video scenes from text or images","Professional image editing with natural language instructions","Rapid video animation from static images with audio","Building production workflows for AI-powered content creation","Automating content generation for marketing and branding","FLAQ AI PTE. LTD.","2026-08-12T18:20:20.487Z",{"last_verified_at":1688,"provider":1689,"adapter":1689,"status":1690,"key_harvested":1691,"api_available_verified":1691,"card_required":429,"card_gate":1692,"card_evidence":1693,"key_test_status":48,"sample_ok":48,"free_tier_summary":48,"what_can_you_do":1694,"limits":48,"card_required_guess":48},"2026-08-12T00:00:21.415Z","probe","card_required",false,"card_text","credit card",[],[1696,1697,1698,1699],"Fal.ai","Replicate","Higgsfield AI","Flyne AI",{"checked_at":1686,"ready":429,"failed":1701},[],"```markdown\n# **Flaq AI – zunifikowana platforma AI do audio, wideo, obrazów i LLM**\n\n**Flaq AI** to wszechstronna platforma API, która łączy modele generatywne (obraz, wideo, audio i LLM) w jednym miejscu, umożliwiając programistom, twórcom i firmom szybką integrację zaawansowanych narzędzi AI. Dzięki modelowi rozliczeń opartemu na kredytach i intuicyjnym narzędziom przeglądarkowym, Flaq AI sprawdzi się zarówno w profesjonalnych projektach, jak i eksperymentach kreatywnych.\n\n---\n\n## **Najważniejsze funkcje**\n\n1. **Zunifikowane API dla wielu modeli**\n   - Jedno spójne API do obsługi modeli **obrazowych** (generacja, edycja), **wideo** (upscaling, animacja), **audio** (syntetyzacja, edycja) oraz **LLM** (rozumowanie, kodowanie, pisanie, analiza plików).\n   - Eliminuje potrzebę korzystania z wielu oddzielnych interfejsów (np. MidJourney + ElevenLabs + OpenAI).\n\n2. **Browserowe narzędzia kreatywne**\n   - Darmowe narzędzia do **testowania modeli bezpośrednio w przeglądarce** (np. generowanie obrazów, edycja w czasie rzeczywistym).\n   - Przydatne dla użytkowników bez doświadczenia programistycznego.\n\n3. **Model Market & Dokumentacja**\n   - Przeglądanie dostępnych modeli z opisami, przykładami i **przewodnikami integracyjnymi**.\n   - Dokumentacja API z przykładami kodu (Python, cURL) ułatwia wdrożenie.\n\n4. **Rozliczenie na kredyty (pay-as-you-go)**\n   - Płatność tylko za **rzeczywiste zużycie** (np. $0.0050\u002Fsek. dla Video Upscaler API, stawki per obraz\u002Fsekunda\u002F1M tokenów).\n   - Przejrzysty panel z saldem i historią zużycia.\n\n5. **Darmowe narzędzia do testów**\n   - Bezpłatny dostęp do podstawowych funkcji generowania i edycji obrazów (bez konieczności podłączania API).\n\n---\n\n## **Dla kogo jest Flaq AI?**\n\n- **Deweloperzy i zespoły IT** – do integracji wielu modeli AI w jednym projekcie (np. aplikacje mobilne, SaaS).\n- **Twórcy treści** – szybka generacja wideo, obrazów lub audio bez skomplikowanych narzędzi.\n- **Agencje marketingowe** – automatyzacja tworzenia materiałów wizualnych i tekstowych.\n- **Badacze i studenci** – testowanie modeli AI w celach edukacyjnych lub badawczych.\n- **Startupy** – prototypowanie produktów z wykorzystaniem AI bez dużych kosztów wstępnych.\n\n---\n\n## **Cena**\n\nFlaq AI działa w modelu **usage-based** (płatność za zużycie), z rozliczeniem w **kredytach**. Przykładowe stawki (zgodnie z danymi ze strony):\n\n- **Video Upscaler API**: od **$0.0050\u002Fsekunda**.\n- **Modele obrazowe\u002Ftekstowe**: ceny per obraz lub per 1M tokenów (dokładne stawki zależą od modelu).\n- **Darmowy plan**:\n  - Dostęp do **browserowych narzędzi** (generowanie\u002Fedycja obrazów).\n  - **Free trial experience** – możliwość przetestowania API bez zobowiązań (saldo startowe na koncie).\n\n*Uwaga*: Dokładne ceny dla poszczególnych modeli są dostępne w [panelu pricing](https:\u002F\u002Fflaq.ai) (mogą ulec zmianie).\n\n---\n\n## **Zalety i wady**\n\n### ✅ **Zalety**\n1. **Wszystko w jednym miejscu** – eliminuje potrzebę korzystania z wielu oddzielnych API (oszczędność czasu i kosztów integracji).\n2. **Elastyczny model płatności** – płacisz tylko za to, czego używasz (idealne dla małych projektów lub testów).\n3. **Przyjazne dla początkujących** – darmowe narzędzia przeglądarkowe i dobra dokumentacja ułatwiają start.\n\n### ❌ **Wady**\n1. **Brak jawnych pakietów abonamentowych** – model pay-as-you-go może być droższy przy dużym wolumenie (np. dla firm z wysokim zużyciem).\n2. **Ograniczone informacje o limitach** – nie wszystkie stawki i limity są jasno wyeksponowane na stronie (wymaga sprawdzenia w panelu).\n3. **Konkurencja cenowa** – niektóre specjalistyczne narzędzia (np. tylko do audio) mogą oferować niższe ceny za pojedyncze usługi.\n\n---\n\n## **Werdykt**\n\n**Flaq AI to solidne rozwiązanie dla tych, którzy potrzebują wszechstronnej platformy AI z jednolitym API i elastycznym modelem płatności.** Sprawdzi się szczególnie w projektach wymagających łączenia wielu modeli (obraz + wideo + tekst) lub szybkiego prototypowania. Darmowe narzędzia przeglądarkowe i trial ułatwiają rozpoczęcie, ale przy dużym wolumenie warto porównać koszty z konkurencją. **Jeśli szukasz \"wszystko-w-jednym\" bez skomplikowanych integracji – Flaq AI jest wart uwagi.**\n```",{"source_page_url":799,"final_url":799,"fetched_at":892},{"alternatives_count":184,"active_deals":184},[],[],{"data":1708},[1709,1719,1725,1733,1741,1748,1755,1762,1770,1777,1784,1791,1798,1806,1813,1819,1826,1834,1842,1850,1857,1864,1872,1879,1887,1894,1900,1907,1914,1921,1928,1935,1942,1948,1955,1962,1969,1975,1980,1985,1990,1994,1999,2004,2009,2013,2018,2022,2027,2031,2035,2039,2043,2047,2051,2054,2058,2062,2066,2069,2073,2076,2079,2083,2086,2089,2092,2095,2098,2101,2104,2108,2111,2114,2117,2121,2130,2135,2138,2143,2151,2155,2159,2163,2167,2175,2183,2187,2191,2195,2199,2202,2209,2216,2224,2232,2236,2240,2248,2252],{"id":1710,"type":1711,"title":1712,"body":1713,"tool_name":1714,"z2u_price":1715,"z2u_url":1716,"status":1717,"read_at":48,"created_at":1718},1348,"price_drop","💰 Vora — spadek ceny","Platforma: Dealify | Kategoria: Sales & Marketing | Cena: €67.95 (regular: €670.95) | Rabat: 90%","Vora",67.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fvora","unread","2026-10-23T04:15:26.716Z",{"id":1720,"type":1711,"title":1721,"body":1713,"tool_name":1722,"z2u_price":1715,"z2u_url":1723,"status":1717,"read_at":48,"created_at":1724},1347,"💰 Seamless QR Code — spadek ceny","Seamless QR Code","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fseamless-qr-code","2026-10-23T04:15:26.351Z",{"id":1726,"type":1711,"title":1727,"body":1728,"tool_name":1729,"z2u_price":1730,"z2u_url":1731,"status":1717,"read_at":48,"created_at":1732},1346,"💰 Inspire Sales Academy — spadek ceny","Platforma: Dealify | Kategoria: Sales & Marketing | Cena: €226.95 (regular: €566.95) | Rabat: 60%","Inspire Sales Academy",226.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Finspire-sales-academy","2026-10-23T04:15:25.498Z",{"id":1734,"type":1711,"title":1735,"body":1736,"tool_name":1737,"z2u_price":1738,"z2u_url":1739,"status":1717,"read_at":48,"created_at":1740},1345,"💰 Contact Concert — spadek ceny","Platforma: Dealify | Kategoria: Sales & Marketing | Cena: €101.95 (regular: €1010.95) | Rabat: 90%","Contact Concert",101.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fcontact-concert","2026-10-23T04:15:24.932Z",{"id":1742,"type":1711,"title":1743,"body":1744,"tool_name":1745,"z2u_price":1715,"z2u_url":1746,"status":1717,"read_at":48,"created_at":1747},1344,"💰 OnChat — spadek ceny","Platforma: Dealify | Kategoria: E-commerce | Cena: €67.95 (regular: €670.95) | Rabat: 90%","OnChat","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fonchat","2026-10-23T04:15:23.130Z",{"id":1749,"type":1711,"title":1750,"body":1751,"tool_name":1752,"z2u_price":1715,"z2u_url":1753,"status":1717,"read_at":48,"created_at":1754},1343,"💰 Helpmate — spadek ceny","Platforma: Dealify | Kategoria: E-commerce | Cena: €67.95 (regular: €101.95) | Rabat: 33%","Helpmate","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fhelpmate","2026-10-23T04:15:22.879Z",{"id":1756,"type":1711,"title":1757,"body":1758,"tool_name":1759,"z2u_price":1715,"z2u_url":1760,"status":1717,"read_at":48,"created_at":1761},1342,"💰 Playix.cloud — spadek ceny","Platforma: Dealify | Kategoria: Cloud | Cena: €67.95 (regular: €670.95) | Rabat: 90%","Playix.cloud","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fplayix-cloud","2026-10-23T04:15:21.439Z",{"id":1763,"type":1711,"title":1764,"body":1765,"tool_name":1766,"z2u_price":1767,"z2u_url":1768,"status":1717,"read_at":48,"created_at":1769},1341,"💰 FileLu — spadek ceny","Platforma: Dealify | Kategoria: Cloud | Cena: €169.95 (regular: €397.95) | Rabat: 57%","FileLu",169.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Ffilelu","2026-10-23T04:15:21.280Z",{"id":1771,"type":1711,"title":1772,"body":1773,"tool_name":1774,"z2u_price":1715,"z2u_url":1775,"status":1717,"read_at":48,"created_at":1776},1340,"💰 Purple Photo — spadek ceny","Platforma: Dealify | Kategoria: Creative | Cena: €67.95 (regular: €89.95) | Rabat: 24%","Purple Photo","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fpurple-photo","2026-10-23T04:15:19.971Z",{"id":1778,"type":1711,"title":1779,"body":1780,"tool_name":1781,"z2u_price":1715,"z2u_url":1782,"status":1717,"read_at":48,"created_at":1783},1339,"💰 CursorClip — spadek ceny","Platforma: Dealify | Kategoria: Creative | Cena: €67.95 (regular: €110.95) | Rabat: 39%","CursorClip","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fcursorclip","2026-10-23T04:15:19.613Z",{"id":1785,"type":1711,"title":1786,"body":1787,"tool_name":1788,"z2u_price":1715,"z2u_url":1789,"status":1717,"read_at":48,"created_at":1790},1338,"💰 Creatiyo — spadek ceny","Platforma: Dealify | Kategoria: Creative | Cena: €67.95 (regular: €226.95) | Rabat: 70%","Creatiyo","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fcreatiyo","2026-10-23T04:15:19.453Z",{"id":1792,"type":1711,"title":1793,"body":1794,"tool_name":1795,"z2u_price":1715,"z2u_url":1796,"status":1717,"read_at":48,"created_at":1797},1337,"💰 Blupry Bundle — spadek ceny","Platforma: Dealify | Kategoria: Creative | Cena: €67.95 (regular: €257.95) | Rabat: 74%","Blupry Bundle","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fblupry","2026-10-23T04:15:19.294Z",{"id":1799,"type":1711,"title":1800,"body":1801,"tool_name":1802,"z2u_price":1803,"z2u_url":1804,"status":1717,"read_at":48,"created_at":1805},1336,"💰 Al Webcam Effects — spadek ceny","Platforma: Dealify | Kategoria: Creative | Cena: €135.95 (regular: €1134.95) | Rabat: 88%","Al Webcam Effects",135.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fal-webcam-effects","2026-10-23T04:15:19.132Z",{"id":1807,"type":1711,"title":1808,"body":1809,"tool_name":1810,"z2u_price":1715,"z2u_url":1811,"status":1717,"read_at":48,"created_at":1812},1335,"💰 Vocal — spadek ceny","Platforma: Dealify | Kategoria: Productivity | Cena: €67.95 (regular: €670.95) | Rabat: 90%","Vocal","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fvocal","2026-10-23T04:15:17.568Z",{"id":1814,"type":1711,"title":1815,"body":1809,"tool_name":1816,"z2u_price":1715,"z2u_url":1817,"status":1717,"read_at":48,"created_at":1818},1334,"💰 Vimageo — spadek ceny","Vimageo","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fvimageo","2026-10-23T04:15:17.350Z",{"id":1820,"type":1711,"title":1821,"body":1822,"tool_name":1823,"z2u_price":1738,"z2u_url":1824,"status":1717,"read_at":48,"created_at":1825},1333,"💰 Sheetany — spadek ceny","Platforma: Dealify | Kategoria: Productivity | Cena: €101.95 (regular: €395.95) | Rabat: 74%","Sheetany","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fsheetany","2026-10-23T04:15:16.584Z",{"id":1827,"type":1711,"title":1828,"body":1829,"tool_name":1830,"z2u_price":1831,"z2u_url":1832,"status":1717,"read_at":48,"created_at":1833},1332,"💰 LightPDF — spadek ceny","Platforma: Dealify | Kategoria: Productivity | Cena: €118.95 (regular: €226.95) | Rabat: 48%","LightPDF",118.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Flightpdf","2026-10-23T04:15:15.734Z",{"id":1835,"type":1711,"title":1836,"body":1837,"tool_name":1838,"z2u_price":1839,"z2u_url":1840,"status":1717,"read_at":48,"created_at":1841},1331,"💰 International Open Academy — spadek ceny","Platforma: Dealify | Kategoria: Productivity | Cena: €203.95 (regular: €453.95) | Rabat: 55%","International Open Academy",203.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Finternational-open-academy","2026-10-23T04:15:15.580Z",{"id":1843,"type":1711,"title":1844,"body":1845,"tool_name":1846,"z2u_price":1847,"z2u_url":1848,"status":1717,"read_at":48,"created_at":1849},1330,"💰 4K Bundle — spadek ceny","Platforma: Dealify | Kategoria: Productivity | Cena: €59.95 (regular: €79.95) | Rabat: 25%","4K Bundle",59.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002F4k-bundle","2026-10-23T04:15:14.911Z",{"id":1851,"type":1711,"title":1852,"body":1853,"tool_name":1854,"z2u_price":1767,"z2u_url":1855,"status":1717,"read_at":48,"created_at":1856},1329,"💰 Video Service Desk — spadek ceny","Platforma: Dealify | Kategoria: Business | Cena: €169.95 (regular: €3390.95) | Rabat: 95%","Video Service Desk","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fvideo-service-desk","2026-10-23T04:15:13.262Z",{"id":1858,"type":1711,"title":1859,"body":1860,"tool_name":1861,"z2u_price":1715,"z2u_url":1862,"status":1717,"read_at":48,"created_at":1863},1328,"💰 Classified Billing — spadek ceny","Platforma: Dealify | Kategoria: Business | Cena: €67.95 (regular: €670.95) | Rabat: 90%","Classified Billing","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fclassified-billing","2026-10-23T04:15:13.024Z",{"id":1865,"type":1711,"title":1866,"body":1867,"tool_name":1868,"z2u_price":1869,"z2u_url":1870,"status":1717,"read_at":48,"created_at":1871},1327,"💰 Paddle CRM — spadek ceny","Platforma: Dealify | Kategoria: Business | Cena: €337.95 (regular: €3373.95) | Rabat: 90%","Paddle CRM",337.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fpaddle-crm","2026-10-23T04:15:12.487Z",{"id":1873,"type":1711,"title":1874,"body":1875,"tool_name":1876,"z2u_price":1738,"z2u_url":1877,"status":1717,"read_at":48,"created_at":1878},1326,"💰 ZuckerBot — spadek ceny","Platforma: Dealify | Kategoria: Social Media | Cena: €101.95 (regular: €1010.95) | Rabat: 90%","ZuckerBot","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fzuckerbot","2026-10-23T04:15:10.775Z",{"id":1880,"type":1711,"title":1881,"body":1882,"tool_name":1883,"z2u_price":1884,"z2u_url":1885,"status":1717,"read_at":48,"created_at":1886},1325,"💰 Sociosight — spadek ceny","Platforma: Dealify | Kategoria: Social Media | Cena: €76.95 (regular: €107.95) | Rabat: 29%","Sociosight",76.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fsociosight","2026-10-23T04:15:10.636Z",{"id":1888,"type":1711,"title":1889,"body":1890,"tool_name":1891,"z2u_price":1715,"z2u_url":1892,"status":1717,"read_at":48,"created_at":1893},1324,"💰 Qura Al — spadek ceny","Platforma: Dealify | Kategoria: Social Media | Cena: €67.95 (regular: €670.95) | Rabat: 90%","Qura Al","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fqura-al","2026-10-23T04:15:10.498Z",{"id":1895,"type":1711,"title":1896,"body":1890,"tool_name":1897,"z2u_price":1715,"z2u_url":1898,"status":1717,"read_at":48,"created_at":1899},1323,"💰 Brand2Social — spadek ceny","Brand2Social","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fbrand2social","2026-10-23T04:15:10.248Z",{"id":1901,"type":1711,"title":1902,"body":1903,"tool_name":1904,"z2u_price":1767,"z2u_url":1905,"status":1717,"read_at":48,"created_at":1906},1322,"💰 Structa — spadek ceny","Platforma: Dealify | Kategoria: Developer Tools | Cena: €169.95 (regular: €339.95) | Rabat: 50%","Structa","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fstructa","2026-10-23T04:15:08.863Z",{"id":1908,"type":1711,"title":1909,"body":1910,"tool_name":1911,"z2u_price":1767,"z2u_url":1912,"status":1717,"read_at":48,"created_at":1913},1321,"💰 N8Nitro — spadek ceny","Platforma: Dealify | Kategoria: Developer Tools | Cena: €169.95 (regular: €1362.95) | Rabat: 88%","N8Nitro","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fn8nitro","2026-10-23T04:15:08.608Z",{"id":1915,"type":1711,"title":1916,"body":1917,"tool_name":1918,"z2u_price":1715,"z2u_url":1919,"status":1717,"read_at":48,"created_at":1920},1320,"💰 EasyLink Automations — spadek ceny","Platforma: Dealify | Kategoria: Developer Tools | Cena: €67.95 (regular: €670.95) | Rabat: 90%","EasyLink Automations","https:\u002F\u002Fdealify.com\u002Fproducts\u002Feasylink-automations","2026-10-23T04:15:08.256Z",{"id":1922,"type":1711,"title":1923,"body":1924,"tool_name":1925,"z2u_price":1730,"z2u_url":1926,"status":1717,"read_at":48,"created_at":1927},1319,"💰 devActivity — spadek ceny","Platforma: Dealify | Kategoria: Developer Tools | Cena: €226.95 (regular: €1124.95) | Rabat: 80%","devActivity","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fdevactivity","2026-10-23T04:15:07.997Z",{"id":1929,"type":1711,"title":1930,"body":1931,"tool_name":1932,"z2u_price":1715,"z2u_url":1933,"status":1717,"read_at":48,"created_at":1934},1318,"💰 WP Website Speedy — spadek ceny","Platforma: Dealify | Kategoria: SEO | Cena: €67.95 (regular: €670.95) | Rabat: 90%","WP Website Speedy","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fwebsite-speedy","2026-10-23T04:15:06.626Z",{"id":1936,"type":1711,"title":1937,"body":1938,"tool_name":1939,"z2u_price":1715,"z2u_url":1940,"status":1717,"read_at":48,"created_at":1941},1317,"💰 LocalAuditPro — spadek ceny","Platforma: Dealify | Kategoria: SEO | Cena: €67.95 (regular: €363.95) | Rabat: 81%","LocalAuditPro","https:\u002F\u002Fdealify.com\u002Fproducts\u002Flocalauditpro","2026-10-23T04:15:06.286Z",{"id":1943,"type":1711,"title":1944,"body":1931,"tool_name":1945,"z2u_price":1715,"z2u_url":1946,"status":1717,"read_at":48,"created_at":1947},1316,"💰 360contentOPS — spadek ceny","360contentOPS","https:\u002F\u002Fdealify.com\u002Fproducts\u002F360contentops","2026-10-23T04:15:05.741Z",{"id":1949,"type":1711,"title":1950,"body":1951,"tool_name":1952,"z2u_price":1715,"z2u_url":1953,"status":1717,"read_at":48,"created_at":1954},1315,"💰 Zenith Write — spadek ceny","Platforma: Dealify | Kategoria: AI | Cena: €67.95 (regular: €670.95) | Rabat: 90%","Zenith Write","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fzenith-write","2026-10-23T04:15:04.293Z",{"id":1956,"type":1711,"title":1957,"body":1958,"tool_name":1959,"z2u_price":1715,"z2u_url":1960,"status":1717,"read_at":48,"created_at":1961},1314,"💰 AI Mentions — spadek ceny","Platforma: Dealify | Kategoria: AI | Cena: €67.95 (regular: €112.95) | Rabat: 40%","AI Mentions","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fai-mentions","2026-10-23T04:15:03.535Z",{"id":1963,"type":1711,"title":1964,"body":1965,"tool_name":1966,"z2u_price":1715,"z2u_url":1967,"status":1717,"read_at":48,"created_at":1968},1313,"💰 Open eLMS Learning Generator — spadek ceny","Platforma: Dealify | Kategoria: AI | Cena: €67.95 (regular: €288.95) | Rabat: 76%","Open eLMS Learning Generator","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fopen-elms","2026-10-23T04:15:03.085Z",{"id":1970,"type":1711,"title":1971,"body":1951,"tool_name":1972,"z2u_price":1715,"z2u_url":1973,"status":1717,"read_at":48,"created_at":1974},1312,"💰 Agentica — spadek ceny","Agentica","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fagentica","2026-10-23T04:15:02.928Z",{"id":1976,"type":1711,"title":1735,"body":1977,"tool_name":1737,"z2u_price":1978,"z2u_url":1739,"status":1717,"read_at":48,"created_at":1979},1311,"Platforma: Dealify | Kategoria: Sales & Marketing | Cena: €100.95 (regular: €1008.95) | Rabat: 90%",100.95,"2026-10-22T04:15:24.021Z",{"id":1981,"type":1711,"title":1764,"body":1982,"tool_name":1766,"z2u_price":1983,"z2u_url":1768,"status":1717,"read_at":48,"created_at":1984},1310,"Platforma: Dealify | Kategoria: Cloud | Cena: €168.95 (regular: €396.95) | Rabat: 57%",168.95,"2026-10-22T04:15:20.592Z",{"id":1986,"type":1711,"title":1800,"body":1987,"tool_name":1802,"z2u_price":1988,"z2u_url":1804,"status":1717,"read_at":48,"created_at":1989},1309,"Platforma: Dealify | Kategoria: Creative | Cena: €134.95 (regular: €1132.95) | Rabat: 88%",134.95,"2026-10-22T04:15:18.661Z",{"id":1991,"type":1711,"title":1821,"body":1992,"tool_name":1823,"z2u_price":1978,"z2u_url":1824,"status":1717,"read_at":48,"created_at":1993},1308,"Platforma: Dealify | Kategoria: Productivity | Cena: €100.95 (regular: €394.95) | Rabat: 74%","2026-10-22T04:15:16.304Z",{"id":1995,"type":1711,"title":1828,"body":1996,"tool_name":1830,"z2u_price":1997,"z2u_url":1832,"status":1717,"read_at":48,"created_at":1998},1307,"Platforma: Dealify | Kategoria: Productivity | Cena: €117.95 (regular: €225.95) | Rabat: 48%",117.95,"2026-10-22T04:15:15.389Z",{"id":2000,"type":1711,"title":1836,"body":2001,"tool_name":1838,"z2u_price":2002,"z2u_url":1840,"status":1717,"read_at":48,"created_at":2003},1306,"Platforma: Dealify | Kategoria: Productivity | Cena: €202.95 (regular: €452.95) | Rabat: 55%",202.95,"2026-10-22T04:15:15.251Z",{"id":2005,"type":1711,"title":1844,"body":2006,"tool_name":1846,"z2u_price":2007,"z2u_url":1848,"status":1717,"read_at":48,"created_at":2008},1305,"Platforma: Dealify | Kategoria: Productivity | Cena: €58.95 (regular: €79.95) | Rabat: 26%",58.95,"2026-10-22T04:15:14.605Z",{"id":2010,"type":1711,"title":1852,"body":2011,"tool_name":1854,"z2u_price":1983,"z2u_url":1855,"status":1717,"read_at":48,"created_at":2012},1304,"Platforma: Dealify | Kategoria: Business | Cena: €168.95 (regular: €3382.95) | Rabat: 95%","2026-10-22T04:15:12.930Z",{"id":2014,"type":1711,"title":1866,"body":2015,"tool_name":1868,"z2u_price":2016,"z2u_url":1870,"status":1717,"read_at":48,"created_at":2017},1303,"Platforma: Dealify | Kategoria: Business | Cena: €336.95 (regular: €3365.95) | Rabat: 90%",336.95,"2026-10-22T04:15:12.188Z",{"id":2019,"type":1711,"title":1874,"body":2020,"tool_name":1876,"z2u_price":1978,"z2u_url":1877,"status":1717,"read_at":48,"created_at":2021},1302,"Platforma: Dealify | Kategoria: Social Media | Cena: €100.95 (regular: €1008.95) | Rabat: 90%","2026-10-22T04:15:10.535Z",{"id":2023,"type":1711,"title":1881,"body":2024,"tool_name":1883,"z2u_price":2025,"z2u_url":1885,"status":1717,"read_at":48,"created_at":2026},1301,"Platforma: Dealify | Kategoria: Social Media | Cena: €75.95 (regular: €107.95) | Rabat: 30%",75.95,"2026-10-22T04:15:10.400Z",{"id":2028,"type":1711,"title":1902,"body":2029,"tool_name":1904,"z2u_price":1983,"z2u_url":1905,"status":1717,"read_at":48,"created_at":2030},1300,"Platforma: Dealify | Kategoria: Developer Tools | Cena: €168.95 (regular: €338.95) | Rabat: 50%","2026-10-22T04:15:08.769Z",{"id":2032,"type":1711,"title":1909,"body":2033,"tool_name":1911,"z2u_price":1983,"z2u_url":1912,"status":1717,"read_at":48,"created_at":2034},1299,"Platforma: Dealify | Kategoria: Developer Tools | Cena: €168.95 (regular: €1359.95) | Rabat: 88%","2026-10-22T04:15:08.509Z",{"id":2036,"type":1711,"title":1866,"body":2037,"tool_name":1868,"z2u_price":1869,"z2u_url":1870,"status":1717,"read_at":48,"created_at":2038},1298,"Platforma: Dealify | Kategoria: Business | Cena: €337.95 (regular: €3369.95) | Rabat: 90%","2026-10-21T04:15:12.016Z",{"id":2040,"type":1711,"title":1735,"body":2041,"tool_name":1737,"z2u_price":1738,"z2u_url":1739,"status":1717,"read_at":48,"created_at":2042},1297,"Platforma: Dealify | Kategoria: Sales & Marketing | Cena: €101.95 (regular: €1009.95) | Rabat: 90%","2026-10-20T04:15:22.708Z",{"id":2044,"type":1711,"title":1821,"body":2045,"tool_name":1823,"z2u_price":1738,"z2u_url":1824,"status":1717,"read_at":48,"created_at":2046},1296,"Platforma: Dealify | Kategoria: Productivity | Cena: €101.95 (regular: €394.95) | Rabat: 74%","2026-10-20T04:15:15.141Z",{"id":2048,"type":1711,"title":1874,"body":2049,"tool_name":1876,"z2u_price":1738,"z2u_url":1877,"status":1717,"read_at":48,"created_at":2050},1295,"Platforma: Dealify | Kategoria: Social Media | Cena: €101.95 (regular: €1009.95) | Rabat: 90%","2026-10-20T04:15:09.607Z",{"id":2052,"type":1711,"title":1764,"body":1765,"tool_name":1766,"z2u_price":1767,"z2u_url":1768,"status":1717,"read_at":48,"created_at":2053},1294,"2026-10-19T04:15:20.999Z",{"id":2055,"type":1711,"title":1800,"body":2056,"tool_name":1802,"z2u_price":1803,"z2u_url":1804,"status":1717,"read_at":48,"created_at":2057},1293,"Platforma: Dealify | Kategoria: Creative | Cena: €135.95 (regular: €1133.95) | Rabat: 88%","2026-10-19T04:15:19.006Z",{"id":2059,"type":1711,"title":1828,"body":2060,"tool_name":1830,"z2u_price":1831,"z2u_url":1832,"status":1717,"read_at":48,"created_at":2061},1292,"Platforma: Dealify | Kategoria: Productivity | Cena: €118.95 (regular: €225.95) | Rabat: 47%","2026-10-19T04:15:15.822Z",{"id":2063,"type":1711,"title":1836,"body":2064,"tool_name":1838,"z2u_price":1839,"z2u_url":1840,"status":1717,"read_at":48,"created_at":2065},1291,"Platforma: Dealify | Kategoria: Productivity | Cena: €203.95 (regular: €452.95) | Rabat: 55%","2026-10-19T04:15:15.663Z",{"id":2067,"type":1711,"title":1844,"body":1845,"tool_name":1846,"z2u_price":1847,"z2u_url":1848,"status":1717,"read_at":48,"created_at":2068},1290,"2026-10-19T04:15:14.981Z",{"id":2070,"type":1711,"title":1852,"body":2071,"tool_name":1854,"z2u_price":1767,"z2u_url":1855,"status":1717,"read_at":48,"created_at":2072},1289,"Platforma: Dealify | Kategoria: Business | Cena: €169.95 (regular: €3385.95) | Rabat: 95%","2026-10-19T04:15:13.154Z",{"id":2074,"type":1711,"title":1881,"body":1882,"tool_name":1883,"z2u_price":1884,"z2u_url":1885,"status":1717,"read_at":48,"created_at":2075},1288,"2026-10-19T04:15:10.578Z",{"id":2077,"type":1711,"title":1902,"body":1903,"tool_name":1904,"z2u_price":1767,"z2u_url":1905,"status":1717,"read_at":48,"created_at":2078},1287,"2026-10-19T04:15:08.899Z",{"id":2080,"type":1711,"title":1909,"body":2081,"tool_name":1911,"z2u_price":1767,"z2u_url":1912,"status":1717,"read_at":48,"created_at":2082},1286,"Platforma: Dealify | Kategoria: Developer Tools | Cena: €169.95 (regular: €1361.95) | Rabat: 88%","2026-10-19T04:15:08.611Z",{"id":2084,"type":1711,"title":1735,"body":1977,"tool_name":1737,"z2u_price":1978,"z2u_url":1739,"status":1717,"read_at":48,"created_at":2085},1285,"2026-10-16T04:15:23.740Z",{"id":2087,"type":1711,"title":1764,"body":1982,"tool_name":1766,"z2u_price":1983,"z2u_url":1768,"status":1717,"read_at":48,"created_at":2088},1284,"2026-10-16T04:15:20.328Z",{"id":2090,"type":1711,"title":1800,"body":1987,"tool_name":1802,"z2u_price":1988,"z2u_url":1804,"status":1717,"read_at":48,"created_at":2091},1283,"2026-10-16T04:15:18.430Z",{"id":2093,"type":1711,"title":1821,"body":1992,"tool_name":1823,"z2u_price":1978,"z2u_url":1824,"status":1717,"read_at":48,"created_at":2094},1282,"2026-10-16T04:15:16.109Z",{"id":2096,"type":1711,"title":1828,"body":1996,"tool_name":1830,"z2u_price":1997,"z2u_url":1832,"status":1717,"read_at":48,"created_at":2097},1281,"2026-10-16T04:15:15.237Z",{"id":2099,"type":1711,"title":1836,"body":2001,"tool_name":1838,"z2u_price":2002,"z2u_url":1840,"status":1717,"read_at":48,"created_at":2100},1280,"2026-10-16T04:15:15.087Z",{"id":2102,"type":1711,"title":1844,"body":2006,"tool_name":1846,"z2u_price":2007,"z2u_url":1848,"status":1717,"read_at":48,"created_at":2103},1279,"2026-10-16T04:15:14.428Z",{"id":2105,"type":1711,"title":1852,"body":2106,"tool_name":1854,"z2u_price":1983,"z2u_url":1855,"status":1717,"read_at":48,"created_at":2107},1278,"Platforma: Dealify | Kategoria: Business | Cena: €168.95 (regular: €3383.95) | Rabat: 95%","2026-10-16T04:15:12.781Z",{"id":2109,"type":1711,"title":1874,"body":2020,"tool_name":1876,"z2u_price":1978,"z2u_url":1877,"status":1717,"read_at":48,"created_at":2110},1277,"2026-10-16T04:15:10.445Z",{"id":2112,"type":1711,"title":1881,"body":2024,"tool_name":1883,"z2u_price":2025,"z2u_url":1885,"status":1717,"read_at":48,"created_at":2113},1276,"2026-10-16T04:15:10.305Z",{"id":2115,"type":1711,"title":1902,"body":2029,"tool_name":1904,"z2u_price":1983,"z2u_url":1905,"status":1717,"read_at":48,"created_at":2116},1275,"2026-10-16T04:15:08.597Z",{"id":2118,"type":1711,"title":1909,"body":2119,"tool_name":1911,"z2u_price":1983,"z2u_url":1912,"status":1717,"read_at":48,"created_at":2120},1274,"Platforma: Dealify | Kategoria: Developer Tools | Cena: €168.95 (regular: €1360.95) | Rabat: 88%","2026-10-16T04:15:08.329Z",{"id":2122,"type":2123,"title":2124,"body":2125,"tool_name":2126,"z2u_price":2127,"z2u_url":2128,"status":1717,"read_at":48,"created_at":2129},1273,"new_tool_found","🆕 PromptBuilder — €44.95 (-93%)","Platforma: Dealify | Kategoria: AI | Cena: €44.95 (regular: €680.95) | Rabat: 93%","PromptBuilder",44.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fpromptbuilder","2026-10-16T04:15:04.054Z",{"id":2131,"type":1711,"title":1712,"body":2132,"tool_name":1714,"z2u_price":2133,"z2u_url":1716,"status":1717,"read_at":48,"created_at":2134},1272,"Platforma: Dealify | Kategoria: Sales & Marketing | Cena: €66.95 (regular: €669.95) | Rabat: 90%",66.95,"2026-10-15T04:15:25.768Z",{"id":2136,"type":1711,"title":1721,"body":2132,"tool_name":1722,"z2u_price":2133,"z2u_url":1723,"status":1717,"read_at":48,"created_at":2137},1271,"2026-10-15T04:15:25.363Z",{"id":2139,"type":1711,"title":1727,"body":2140,"tool_name":1729,"z2u_price":2141,"z2u_url":1731,"status":1717,"read_at":48,"created_at":2142},1270,"Platforma: Dealify | Kategoria: Sales & Marketing | Cena: €225.95 (regular: €566.95) | Rabat: 60%",225.95,"2026-10-15T04:15:24.523Z",{"id":2144,"type":1711,"title":2145,"body":2146,"tool_name":2147,"z2u_price":2148,"z2u_url":2149,"status":1717,"read_at":48,"created_at":2150},1269,"💰 Glanc AI — spadek ceny","Platforma: Dealify | Kategoria: Sales & Marketing | Cena: €274.95 (regular: €782.95) | Rabat: 65%","Glanc AI",274.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fglanc-ai","2026-10-15T04:15:24.274Z",{"id":2152,"type":1711,"title":1743,"body":2153,"tool_name":1745,"z2u_price":2133,"z2u_url":1746,"status":1717,"read_at":48,"created_at":2154},1268,"Platforma: Dealify | Kategoria: E-commerce | Cena: €66.95 (regular: €669.95) | Rabat: 90%","2026-10-15T04:15:22.062Z",{"id":2156,"type":1711,"title":1750,"body":2157,"tool_name":1752,"z2u_price":2133,"z2u_url":1753,"status":1717,"read_at":48,"created_at":2158},1267,"Platforma: Dealify | Kategoria: E-commerce | Cena: €66.95 (regular: €101.95) | Rabat: 34%","2026-10-15T04:15:21.821Z",{"id":2160,"type":1711,"title":1757,"body":2161,"tool_name":1759,"z2u_price":2133,"z2u_url":1760,"status":1717,"read_at":48,"created_at":2162},1266,"Platforma: Dealify | Kategoria: Cloud | Cena: €66.95 (regular: €669.95) | Rabat: 90%","2026-10-15T04:15:20.429Z",{"id":2164,"type":1711,"title":1772,"body":2165,"tool_name":1774,"z2u_price":2133,"z2u_url":1775,"status":1717,"read_at":48,"created_at":2166},1265,"Platforma: Dealify | Kategoria: Creative | Cena: €66.95 (regular: €89.95) | Rabat: 26%","2026-10-15T04:15:19.024Z",{"id":2168,"type":1711,"title":2169,"body":2170,"tool_name":2171,"z2u_price":2172,"z2u_url":2173,"status":1717,"read_at":48,"created_at":2174},1264,"💰 MiPLR — spadek ceny","Platforma: Dealify | Kategoria: Creative | Cena: €32.95 (regular: €89.95) | Rabat: 63%","MiPLR",32.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fmiplr","2026-10-15T04:15:18.871Z",{"id":2176,"type":2123,"title":2177,"body":2178,"tool_name":2179,"z2u_price":2180,"z2u_url":2181,"status":1717,"read_at":48,"created_at":2182},1263,"🆕 KDPage — €78.95 (-65%)","Platforma: Dealify | Kategoria: Creative | Cena: €78.95 (regular: €225.95) | Rabat: 65%","KDPage",78.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fkdpage","2026-10-15T04:15:18.725Z",{"id":2184,"type":1711,"title":1779,"body":2185,"tool_name":1781,"z2u_price":2133,"z2u_url":1782,"status":1717,"read_at":48,"created_at":2186},1262,"Platforma: Dealify | Kategoria: Creative | Cena: €66.95 (regular: €110.95) | Rabat: 40%","2026-10-15T04:15:18.572Z",{"id":2188,"type":1711,"title":1786,"body":2189,"tool_name":1788,"z2u_price":2133,"z2u_url":1789,"status":1717,"read_at":48,"created_at":2190},1261,"Platforma: Dealify | Kategoria: Creative | Cena: €66.95 (regular: €225.95) | Rabat: 70%","2026-10-15T04:15:18.431Z",{"id":2192,"type":1711,"title":1793,"body":2193,"tool_name":1795,"z2u_price":2133,"z2u_url":1796,"status":1717,"read_at":48,"created_at":2194},1260,"Platforma: Dealify | Kategoria: Creative | Cena: €66.95 (regular: €257.95) | Rabat: 74%","2026-10-15T04:15:18.289Z",{"id":2196,"type":1711,"title":1808,"body":2197,"tool_name":1810,"z2u_price":2133,"z2u_url":1811,"status":1717,"read_at":48,"created_at":2198},1259,"Platforma: Dealify | Kategoria: Productivity | Cena: €66.95 (regular: €669.95) | Rabat: 90%","2026-10-15T04:15:16.690Z",{"id":2200,"type":1711,"title":1815,"body":2197,"tool_name":1816,"z2u_price":2133,"z2u_url":1817,"status":1717,"read_at":48,"created_at":2201},1258,"2026-10-15T04:15:16.547Z",{"id":2203,"type":1711,"title":2204,"body":2205,"tool_name":2206,"z2u_price":2172,"z2u_url":2207,"status":1717,"read_at":48,"created_at":2208},1257,"💰 StickyPassword — spadek ceny","Platforma: Dealify | Kategoria: Productivity | Cena: €32.95 (regular: €218.95) | Rabat: 85%","StickyPassword","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fstickypassword","2026-10-15T04:15:15.904Z",{"id":2210,"type":1711,"title":2211,"body":2212,"tool_name":2213,"z2u_price":2172,"z2u_url":2214,"status":1717,"read_at":48,"created_at":2215},1256,"💰 Notepad — spadek ceny","Platforma: Dealify | Kategoria: Productivity | Cena: €32.95 (regular: €329.95) | Rabat: 90%","Notepad","https:\u002F\u002Fdealify.com\u002Fproducts\u002Fnotepad","2026-10-15T04:15:15.108Z",{"id":2217,"type":1711,"title":2218,"body":2219,"tool_name":2220,"z2u_price":2221,"z2u_url":2222,"status":1717,"read_at":48,"created_at":2223},1255,"💰 FlashBooks — spadek ceny","Platforma: Dealify | Kategoria: Productivity | Cena: €282.95 (regular: €1973.95) | Rabat: 86%","FlashBooks",282.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fflashbooks","2026-10-15T04:15:14.664Z",{"id":2225,"type":1711,"title":2226,"body":2227,"tool_name":2228,"z2u_price":2229,"z2u_url":2230,"status":1717,"read_at":48,"created_at":2231},1254,"💰 Envoice — spadek ceny","Platforma: Dealify | Kategoria: Business | Cena: €157.95 (regular: €1576.95) | Rabat: 90%","Envoice",157.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fenvoice","2026-10-15T04:15:12.373Z",{"id":2233,"type":1711,"title":1859,"body":2234,"tool_name":1861,"z2u_price":2133,"z2u_url":1862,"status":1717,"read_at":48,"created_at":2235},1253,"Platforma: Dealify | Kategoria: Business | Cena: €66.95 (regular: €669.95) | Rabat: 90%","2026-10-15T04:15:12.225Z",{"id":2237,"type":1711,"title":1866,"body":2238,"tool_name":1868,"z2u_price":2016,"z2u_url":1870,"status":1717,"read_at":48,"created_at":2239},1252,"Platforma: Dealify | Kategoria: Business | Cena: €336.95 (regular: €3369.95) | Rabat: 90%","2026-10-15T04:15:11.677Z",{"id":2241,"type":1711,"title":2242,"body":2243,"tool_name":2244,"z2u_price":2245,"z2u_url":2246,"status":1717,"read_at":48,"created_at":2247},1251,"💰 LeadLocator — spadek ceny","Platforma: Dealify | Kategoria: Business | Cena: €191.95 (regular: €1917.95) | Rabat: 90%","LeadLocator",191.95,"https:\u002F\u002Fdealify.com\u002Fproducts\u002Fleadlocator","2026-10-15T04:15:11.332Z",{"id":2249,"type":1711,"title":1889,"body":2250,"tool_name":1891,"z2u_price":2133,"z2u_url":1892,"status":1717,"read_at":48,"created_at":2251},1250,"Platforma: Dealify | Kategoria: Social Media | Cena: €66.95 (regular: €669.95) | Rabat: 90%","2026-10-15T04:15:09.802Z",{"id":2253,"type":1711,"title":1896,"body":2250,"tool_name":1897,"z2u_price":2133,"z2u_url":1898,"status":1717,"read_at":48,"created_at":2254},1249,"2026-10-15T04:15:09.556Z",{"data":2256},{"left":305,"right":800,"confidence":2257,"content":2258,"generatedAt":2287,"model":2288},0.65,{"h1":2259,"lead":2260,"sections":2261,"verdict":2274,"faq":2277},"Inkling vs Flaq AI — multimodalne API w praktyce","Porównujemy otwarty model Inkling (Thinking Machines Lab) z platformą API Flaq AI. Sprawdzamy, które rozwiązanie lepiej sprawdzi się w generowaniu treści, a które w integracji z istniejącymi systemami.",[2262,2265,2268,2271],{"h2":2263,"body":2264},"Czym różnią się modele dostępu?","Inkling to otwarty model o wadze 975B parametrów, dostępny na licencji Apache 2.0 — możesz go hostować samodzielnie lub używać przez API partnerów (np. Together AI, Baseten) w cenie 4,05 dolara za 1000 tokenów. Flaq AI to platforma API oferująca dostęp do wielu modeli (np. GPT Image 2, Seedance, Claude Opus 4.7) z cenami od 0,005 dolara za sekundę wideo i 0,008 dolara za obraz. Flaq ma darmowe narzędzia do testowania, ale nie oferuje samodzielnego hostingu.",{"h2":2266,"body":2267},"Jakie są możliwości przetwarzania treści?","Inkling przyjmuje tekst, obrazy i audio, a generuje tekst i kod — ma kontekst do 1 miliona tokenów, co pozwala na analizę bardzo długich dokumentów. Flaq AI koncentruje się na generowaniu obrazów, wideo i tekstu, oferując m.in. generowanie wideo z tekstu lub obrazu w rozdzielczości 768x512 przy 24 FPS. Inkling wspiera wywoływanie narzędzi i strukturalne odpowiedzi, Flaq AI daje kontrolę nad postaciami, scenami i stylem wizualnym.",{"h2":2269,"body":2270},"Które rozwiązanie jest tańsze w praktyce?","Inkling ma zerowy koszt licencji, ale płacisz za inferencję — 4,05 dolara za 1000 tokenów to koszt porównywalny z drogimi modelami API. Flaq AI oferuje darmowe narzędzia do generowania i edycji obrazów, a płatne API zaczyna się od 0,005 dolara za sekundę wideo — przy typowych zastosowaniach może być tańszy. Jednak przy dużych wolumenach tekstu (np. analiza 1M tokenów) koszt Inklinga może szybko wzrosnąć.",{"h2":2272,"body":2273},"Które narzędzie lepiej sprawdzi się w produkcji?","Inkling daje pełną kontrolę dzięki samodzielnemu hostingowi i wielu formatom kwantyzacji (GGUF, BF16, NVFP4) — to dobre dla firm z własną infrastrukturą. Flaq AI to gotowe API z wieloma modelami, co przyspiesza wdrożenie i pozwala łatwo przełączać się między modelami. Flaq nie ma jednak opcji self-hostingu, więc jesteś zależny od dostawcy.",{"pick_x":2275,"pick_y":2276},"Inkling wybierzesz, jeśli potrzebujesz otwartego modelu z ogromnym kontekstem (1M tokenów) i chcesz hostować go samodzielnie — to opcja dla zespołów ML i firm z infrastrukturą.","Flaq AI wybierzesz, jeśli szukasz prostego, taniego API do generowania obrazów i wideo — to dobre dla deweloperów i małych firm bez własnych GPU.",[2278,2281,2284],{"q":2279,"a":2280},"Czy Inkling obsługuje generowanie wideo?","Nie, Inkling generuje tylko tekst i kod. Wideo generuje Flaq AI.",{"q":2282,"a":2283},"Czy Flaq AI ma darmowy trial?","Tak, Flaq AI oferuje darmowe narzędzia do generowania i edycji obrazów.",{"q":2285,"a":2286},"Czy Inkling można używać bezpłatnie?","Wagi modelu są darmowe (Apache 2.0), ale inferencja przez API kosztuje 4,05 dolara za 1000 tokenów.","2026-08-10T23:27:46.678Z","deepseek\u002Fdeepseek-v4-flash"]