Free AI Models & APIs
Every genuinely-free AI model & API (open-weight, free tiers, and no-signup tools), ranked by the community. Each row shows how to use it, how it performs, the free limits, and the catch. No paywalls, no shared keys, no trial-credit traps.
Using these models in production? We build it. Book a call
| # | Model | Modality | Access | Benchmarksas of Nov 11 2026 | Free limits | The catch | ▲ Votes | How to use |
|---|---|---|---|---|---|---|---|---|
| 1 |  Ollama Cloud · No longer free · Checked Nov 11, 2026 | CODETEXT | Free API | No longer free (checked Nov 11, 2026) | ✓ no card · ? comm | |||
| 2 |  Z.ai · Checked Jul 30, 2026 | TEXTCODE | Open-weight | Self-host free | ✓ no card · ✓ commercial | |||
| 3 |  Google Gemini · No longer free · Checked Nov 11, 2026 | IMAGEVIDEOAUDIOTEXT | Free API | No longer free (checked Nov 11, 2026) | ✓ no card · ? comm | |||
| 4 |  DeepSeek · Checked Jul 30, 2026 | TEXTCODE | Open-weight | Self-host free | ✓ no card · ✓ commercial | |||
| 5 |  Google AI Studio · Checked Nov 11, 2026 | TEXTIMAGEAUDIO | Free API | Free tier · existing users only, shuts down Dec 1, 2026 | ✓ no card · ✓ commercial | |||
| 6 | Google Gemini · Checked Nov 11, 2026 | IMAGEVIDEOAUDIOTEXT | Free API | Free tier · existing users only, shuts down Dec 1, 2026 | ✓ no card · ? comm | |||
| 7 | CODETEXT | Free API | no published score | One-time 1M tokens, 90 days (new users) | ✓ no card · ? comm | |||
| 8 |  Scaleway · Checked Nov 12, 2026 | CODETEXT | Free API | First 1M tokens free (shared) · then paid | ✓ no card · ? comm | |||
| 9 |  NVIDIA NIM · Checked Nov 11, 2026 | IMAGEVIDEOAUDIOTEXT | Free API | no published score | Prototyping only · rate limits not published | ✓ no card · ? comm | ||
| 10 |  GitHub Models · No longer free · Checked Nov 11, 2026 | IMAGETEXT | Free API | no published score | No longer free (checked Nov 11, 2026) | ✓ no card · ? comm | ||
| 11 |  Qwen · No longer free · Checked Nov 11, 2026 | CODETEXT | Free API | no published score | No longer free (checked Nov 11, 2026) | ✓ no card · ? comm | ||
| 12 |  Mistral · No longer free · Checked Nov 11, 2026 | CODETEXT | Free API | No longer free (checked Nov 11, 2026) | ✓ no card · ? comm | |||
| 13 |  OpenRouter · Checked Nov 11, 2026 | TEXTCODE | Free API | no published score | 20 RPM · 50 RPD (1,000 RPD after 10+ credits) | ✓ no card · ✓ commercial | ||
| 14 |  Cloudflare Workers AI · Checked Nov 12, 2026 | TEXT | Free API | no published score | 10K neurons/day (shared) | ✓ no card · ? comm | ||
| 15 | Google AI Studio · No longer free · Checked Nov 11, 2026 | CODETEXT | Free API | no published score | No longer free (checked Nov 11, 2026) | ✓ no card · ? comm | ||
| 16 |  Hugging Face · No longer free · Checked Nov 11, 2026 | TEXTIMAGE | Free API | no published score | No longer free (checked Nov 11, 2026) | ✓ no card · ✓ commercial | ||
| 17 |  NVIDIA NIM · Checked Jun 26, 2026 | CODETEXT | Free API | Free tier | ✓ no card · ? comm | |||
| 18 | OpenCode Zen · No longer free · Checked Nov 11, 2026 | CODETEXT | Free API | No longer free (checked Nov 11, 2026) | ✓ no card · ? comm | |||
| 19 | Qwen · Checked Jun 26, 2026 | CODETEXT | Free API | Free tier | ✓ no card · ? comm | |||
| 20 |  Groq · No longer free · Checked Nov 11, 2026 | TEXTCODE | Free API | No longer free (checked Nov 11, 2026) | ✓ no card · ✓ commercial | |||
| 21 |  Pollinations.ai · Checked Jun 26, 2026 | IMAGETEXT | No-signup | no published score | No key · unlimited-ish | ✓ no card · △ non-comm | ||
| 22 | Google Gemini · Checked Nov 11, 2026 | IMAGEVIDEOAUDIOTEXT | Free API | Free tier · existing users only, shuts down Dec 1, 2026 | ✓ no card · ? comm | |||
| 23 | OpenCode Zen · Checked Nov 12, 2026 | CODETEXT | Free API | no published score | Free for a limited time · limits not published | ✓ no card · ? comm | ||
| 24 |  OVHcloud AI Endpoints · Checked Nov 12, 2026 | TEXT | Free API | no published score | 2 RPM per IP (no key) | ✓ no card · ? comm | ||
| 25 | GitHub Models · No longer free · Checked Nov 11, 2026 | CODETEXT | Free API | no published score | No longer free (checked Nov 11, 2026) | ✓ no card · ? comm |
Picked a free model and hit the rate limit?
A senior engineer looks at your setup for 30 minutes. You leave knowing which model fits, what it costs once you outgrow the free tier, and what to build around it. You keep the notes either way.
Show me which model fitsBrowse all 395 free models by provider
Black Forest Labs
Cloudflare Workers AI
- @cf/deepseek-ai/deepseek-r1-distill-qwen-32b
- @cf/google/gemma-4-26b-a4b-it
- @cf/meta/llama-3.3-70b-instruct-fp8-fast
- @cf/meta/llama-4-scout-17b-16e-instruct
- @cf/mistralai/mistral-small-3.1-24b-instruct
- @cf/openai/gpt-oss-120b
- @cf/zhipuai/glm-4.7-flash
- + 42 more models
- apertus-v1.5-8b
- aura-1
- aura-2-en
- aura-2-es
- bge-base-en-v1.5
- bge-large-en-v1.5
- bge-m3
- bge-reranker-base
- bge-small-en-v1.5
- clairic
- clef
- clef-flash
- clef-omni
- distilbert-sst-2-int8
- dreamshaper-8-lcm
- eurollm-9b-it
- flux
- flux-1-schnell
- flux-2-dev
- flux-2-klein-4b
- flux-2-klein-9b
- gemma-sea-lion-v4-27b-it
- GLM-4.7-Flash
- GPT OSS 20B
- Granite 4.0 Micro
- indictrans2-en-indic-1B
- llama-3.1-8b-instruct-fp8
- llama-3.2-11b-vision-instruct
- llama-3.2-1b-instruct
- llama-3.2-3b-instruct
- llama-guard-3-8b
- lucid-origin
- m2m100-1.2b
- melotts
- Mistral 7B v0.2
- mistral-7b-instruct-v0.2-lora
- moondream3.1-9B-A2B
- Nemotron 3 Super
- nova-3
- phoenix-1.0
- phonellm-mini-alpha-2
- plamo-embedding-1b
- Qwen2.5 Coder 32B
- Qwen3 30B MoE
- qwen3-embedding-0.6b
- qwen3.8-27b
- qwq-32b
- resnet-50
- whisper
- whisper-large-v3-turbo
Cohere
DeepSeek
ElevenLabs
Google AI Studio
- Gemini 2.5 Flash
- Gemini 2.5 Flash Lite
- Gemini 2.5 Flash Native Audio Preview
- Gemini 2.5 Flash Preview TTS
- Gemini 2.5 Pro
- Gemini 3 Flash Preview
- Gemini 3.1 Flash Lite
- Gemini 3.1 Flash Live Preview
- Gemini 3.1 Flash TTS Preview
- Gemini 3.5 Flash-Lite
- Gemini 3.5 Live Translate Preview
- Gemini 3.5 Transcribe
- Gemini 3.5 Transcribe Live
- Gemini 3.6 Flash
- Gemini 3.8 Flash
- Gemini 3.8 Flash TTS
- Gemini 3.8 Flash-Lite TTS
- Gemini 3.8 Live
- Gemini 3.8 Live Extended Thinking
- Gemini Decision
- Gemini Embedding 001
- Gemini Embedding 2
- Gemini Robotics ER 2 Preview
- Gemini Robotics ER 2 Streaming Preview
Google Gemini
Groq
Kilo Code
ModelScope
- + API-Inference-enabled models
- AntAngelMed
- CompassJudger-1-32B-Instruct
- DeepSeek-V4-Flash-0731
- DeepSeek-V4-Pro
- DeepSeek-V4-Pro-0813
- DeepSeek-V4.1-Flash
- DeepSeek-V4.1-Flash-1017
- EA-29B-A4B
- ERNIE-4.5-0.3B-PT
- ERNIE-4.5-21B-A3B-PT
- ERNIE-4.5-300B-A47B-PT
- ERNIE-4.5-VL-28B-A3B-PT
- GLM-4.7-Flash
- GLM-5.2
- Intern-S1
- Intern-S1-mini
- Intern-S2-Preview
- InternVL3_5-241B-A28B
- Kimi-K3.1-Pro
- LongCat-Flash-Lite
- MiniMax-M1-80k
- MiniMax-M3
- Mistral-Large-Instruct-2407
- Nex-N2.5-mini
- Nex-N2.5-Pro
- Qwen-Image-Edit
- Qwen-Image-Edit (Qwen)
- Qwen/Qwen3.5-27B
- Qwen/Qwen3.5-35B-A3B
- Qwen3.5-122B-A10B
- Qwen3.5-397B-A17B
- Qwen3.8-27B
- Qwen3.8-Flash-Next
- Step-3.5-Flash
- Step-3.7-Flash
- Step-5-Preview
- XiYanSQL-QwenCoder-32B-2412
- XiYanSQL-QwenCoder-32B-2504
NVIDIA
NVIDIA NIM
- + 90 more models
- Arctic Embed L
- CodeGemma 1.1 7B
- CodeGemma 7B
- CodeLlama 70B
- Codestral 22B Instruct v0.1
- Cosmos Reason2 8B
- DBRX Instruct
- DeepSeek Coder 6.7B Instruct
- DeepSeek V4 Flash
- DeepSeek V4 Pro
- DeepSeek V4.1 Flash
- deepseek-ai/deepseek-r1
- DiffusionGemma 26B A4B IT
- Embed QA 4
- Gemma 2B
- Gemma 3 12B IT
- Gemma 3 4B IT
- Gemma 4 31B
- GLM 5.1
- GLM 5.3 Flash
- google/gemma-4-31b
- GPT OSS 120B
- GPT OSS 20B
- Granite 3.0 3B A800M Instruct
- Granite 3.0 8B Instruct
- Granite 34B Code Instruct
- Granite 8B Code Instruct
- Jamba 1.5 Large Instruct
- Kimi K2.6
- Laguna XS 2.1
- Llama 2 70B
- Llama 3.1 Nemotron 51B Instruct
- Llama 3.1 Nemotron 70B Instruct
- Llama 3.2 11B Vision
- Llama 3.2 90B Vision Instruct
- Llama 3.2 NemoRetriever 1B VLM Embed v1
- Llama 3.2 NV EmbedQA 1B v1
- Llama 4 Maverick
- Llama Nemotron Embed VL 1B v2
- Llama3 ChatQA 1.5 70B
- meta/llama-3.1-405b-instruct
- MiniMax M2.7
- MiniMax M3
- MiniMax M3.1 Flash
- minimax/minimax-m2.7
- Ministral 14B
- Mistral 7B Instruct v0.3
- Mistral Large
- Mistral Large 4 1T A52B
- Mistral Large 675B
- Mistral Medium 3.5
- Mistral NeMo 12B Instruct
- Mistral NeMo Minitron 8B 8K Instruct
- Mistral Small 4
- mistralai/mistral-large-2-instruct
- Mixtral 8x22B v0.1
- Nemotron 3 Embed 1B
- Nemotron 3 Omni
- Nemotron 3 Ultra
- Nemotron 3.5 Bolt E4B
- Nemotron 3.5 Embed 1B
- Nemotron 3.5 Embed 8B
- Nemotron 3.5 Lightning 30B A3B
- Nemotron 3.5 Super VL 120B A12B
- Nemotron Nano 3 30B A3B
- Nemotron-4 340B Instruct
- NV EmbedQA Mistral 7B v2
- nvidia/llama-3.1-nemotron-ultra-253b-v1
- nvidia/nemotron-3-nano-30b-a3b
- nvidia/nemotron-3-super-120b-a12b
- Palmyra Creative 122B
- Palmyra Fin 70B 32K
- Palmyra Med 70B
- Palmyra Med 70B 32K
- Phi 4 Mini
- Phi-3 Vision 128K Instruct
- Phi-3.5 MoE Instruct
- qwen/qwen2.5-72b-instruct
- Qwen3 80B Instruct
- Qwen3.5 122B
- Qwen3.5 400B VLM
- RecurrentGemma 2B
- Riva Translate 4B Instruct
- Riva Translate 4B Instruct v2
- SEA-LION 7B Instruct
- Seed OSS 36B
- StarCoder2 15B
- Step 3.5 Flash
- Step 3.7 Flash
- Step 5 Preview
- Stockmark 100B
- Yi Large
- Zamba2 7B Instruct
OpenAI
OpenCode Zen
OpenRouter
OVHcloud AI Endpoints
- bge-m3
- bge-multilingual-gemma2 (being retired, EOL Jan 21 2027)
- gpt-oss-120b
- gpt-oss-20b
- Meta-Llama-3_3-70B-Instruct
- Mistral-Small-3.2-24B-Instruct
- nvr-tts-de-de
- nvr-tts-en-us
- nvr-tts-es-es
- nvr-tts-it-it
- Qwen2.5-VL-72B-Instruct
- Qwen3-Embedding-8B
- Qwen3.5-397B-A17B
- Qwen3.5-9B
- Qwen3.6-27B
- Qwen3.8-27B
- Qwen3Guard-Gen-0.6B (beta, listed Free)
- Qwen3Guard-Gen-8B (beta, listed Free)
- stable-diffusion-xl-base-v10
- whisper-large-v3
- whisper-large-v3-turbo
Pollinations.ai
Routeway
SambaNova
Scaleway
SiliconFlow
Stability AI
Z AI (Zhipu AI)
Z.ai
Free monthly update
The monthly free-models update
What is newly free, what got rate-limited, and what to switch to. One email a month. Unsubscribe anytime.
Free AI models, honestly answered
Are free AI models really free?
Three kinds of free exist: open-weight models you self-host (free forever, you pay only infrastructure), permanent free API tiers (free within rate limits), and no-signup tools. The catch is usually rate limits, non-commercial clauses, or your data being used for training. Every listing here shows its catch.
What is the best free AI model right now?
It depends on the job: the community votes on this page rank what practitioners actually rate highest, and you can filter by text, code, image, video, audio, or embeddings. For raw quality per dollar, open-weight frontier models you self-host are hard to beat.
Can I use free AI models commercially?
Often yes for permissive open-weight licenses (Apache-2.0, MIT), but many free API tiers prohibit commercial use and some community licenses add revenue caps. Check the commercial column on each listing and verify the current license before shipping.
Do free AI API keys without signup exist?
Yes, a small set of services work with no key at all, and they are labeled No-signup here. Never use shared API keys posted online: they run on someone else's billing, violate provider terms, and get revoked.
How do I use a free model in my own project?
Most free API tiers are OpenAI-compatible, so you point your SDK at the provider base URL with your free key. Every listing has a How drawer with setup steps, a quickstart snippet, and a copy-paste prompt that walks Claude Code through the setup for you.
Free model guides
- Full AI Model Leaderboard — paid and free, benchmarks and speed
- Live AI Model Pricing Index
- Free AI models for coding — Cursor, VS Code, Claude Code (2026)
- Best free AI models for n8n — no-token agent workflows (2026)
- Free AI models for commercial use — license-safe options (2026)
- Best local LLM tools — run models locally in 2026
Wiring one of these into a production workflow instead of a script? Our n8n automation agency builds and maintains the automation around the model.
How we rank & verify. Listings are ordered by community upvotes (one vote per model per device, no login); ties fall back to the LMArena text Elo. The Benchmarks column shows published numbers only, each linked to its source and dated Nov 11 2026: the Artificial Analysis Intelligence Index, LMArena text Elo and SWE-bench Verified. A model gets a number only when the source tested that exact model and version. We never blend them into one score. “Free” claims and limits are re-checked against each provider's own docs or API and dated per row, with a link to the source page. A model that stops being free stays listed as “No longer free” for 7 days, then leaves the list.
Sources. Seeded and cross-checked against mnfst/awesome-free-llm-apis · zebbern/no-cost-ai · 12britz/awesome-free-models · vava-nessa/free-coding-models · open-free-llm-api · cheahjs/free-llm-api-resources. We never list or link shared API keys.