Free LLM Free LLM APIs and their rate limits
Which LLM APIs can you call for free, how many requests and tokens you get per minute and per day, whether you need a card, and what the provider does with your prompts. 13 providers, from Google Gemini, Groq, Cerebras and SambaNova to OpenRouter's :free models, Mistral, Cloudflare and Vercel. Every row links to the official page with the date we checked it. If a provider doesn't publish a number, the cell says n/p (not published).
2026-10-25last checked
Tap a box to filter by it; tap it again to remove that filter. Filters combine.
66 of 66 free models / offers. n/p = not published on the provider's page.
| Model | Provider | Requests | Tokens | Other limits / notes | Card? | Source |
|---|---|---|---|---|---|---|
| gemini-3.8-flash | Gemini Free tier with rate limits | n/p | n/p | See provider page | no | ai.google.dev checked 2026-10-25 |
| gemini-3.6-flash | Gemini Free tier with rate limits | n/p | n/p | See provider page | no | ai.google.dev checked 2026-10-25 |
| gemini-3.5-flash-lite | Gemini Free tier with rate limits | n/p | n/p | See provider page | no | ai.google.dev checked 2026-10-25 |
| gemini-3.1-flash-lite | Gemini Free tier with rate limits | n/p | n/p | See provider page | no | ai.google.dev checked 2026-10-25 |
| gemini-3-flash-preview | Gemini Free tier with rate limits | n/p | n/p | See provider page | no | ai.google.dev checked 2026-10-25 |
| gemini-2.5-pro | Gemini Free tier with rate limits | n/p | n/p | See provider page | no | ai.google.dev checked 2026-10-25 |
| gemini-2.5-flash | Gemini Free tier with rate limits | n/p | n/p | See provider page | no | ai.google.dev checked 2026-10-25 |
| gemini-2.5-flash-lite | Gemini Free tier with rate limits | n/p | n/p | See provider page | no | ai.google.dev checked 2026-10-25 |
| gemini-embedding-2 | Gemini Free tier with rate limits | n/p | n/p | See provider page | no | ai.google.dev checked 2026-10-25 |
| openai/gpt-oss-120b | Groq Free tier with rate limits | 30/min · 1K/day | 8K/min · 200K/day | See provider page | not stated | console.groq.com checked 2026-10-25 |
| openai/gpt-oss-20b | Groq Free tier with rate limits | 30/min · 1K/day | 8K/min · 200K/day | See provider page | not stated | console.groq.com checked 2026-10-25 |
| openai/gpt-oss-safeguard-20b | Groq Free tier with rate limits | 3/min · 1K/day | 2K/min · 200K/day | See provider page | not stated | console.groq.com checked 2026-10-25 |
| qwen/qwen3.8-27b | Groq Free tier with rate limits | 30/min · 1K/day | 8K/min · 200K/day | See provider page | not stated | console.groq.com checked 2026-10-25 |
| whisper-large-v3 | Groq Free tier with rate limits | 20/min · 2K/day | n/p | 7,200 audio seconds/hour, 28,800 audio seconds/day | not stated | console.groq.com checked 2026-10-25 |
| whisper-large-v3-turbo | Groq Free tier with rate limits | 20/min · 2K/day | n/p | 7,200 audio seconds/hour, 28,800 audio seconds/day | not stated | console.groq.com checked 2026-10-25 |
| meta-llama/llama-prompt-guard-2-22m | Groq Free tier with rate limits | 30/min · 14.4K/day | 15K/min · 500K/day | See provider page | not stated | console.groq.com checked 2026-10-25 |
| meta-llama/llama-prompt-guard-2-86m | Groq Free tier with rate limits | 30/min · 14.4K/day | 15K/min · 500K/day | See provider page | not stated | console.groq.com checked 2026-10-25 |
| canopylabs/orpheus-v1-english | Groq Free tier with rate limits | 10/min · 100/day | 1.2K/min · 3.6K/day | See provider page | not stated | console.groq.com checked 2026-10-25 |
| canopylabs/orpheus-arabic-saudi | Groq Free tier with rate limits | 10/min · 100/day | 1.2K/min · 3.6K/day | See provider page | not stated | console.groq.com checked 2026-10-25 |
| gpt-oss-120b | Cerebras Free tier with rate limits | 5/min | 30K/min · 1M/day | 90K total tokens/min, 1M tokens/hour | not stated | inference-docs.cerebras.ai checked 2026-10-25 |
| qwen-3.8-27b | Cerebras Free tier with rate limits | 5/min | 30K/min · 1M/day | 90K total tokens/min, 1M tokens/hour | not stated | inference-docs.cerebras.ai checked 2026-10-25 |
| DeepSeek-V3.1 | SambaNova Free tier with rate limits | 20/min · 20/day | 200K/day | See provider page | no | docs.sambanova.ai checked 2026-10-25 |
| Meta-Llama-3.3-70B-Instruct | SambaNova Free tier with rate limits | 20/min · 20/day | 200K/day | See provider page | no | docs.sambanova.ai checked 2026-10-25 |
| gpt-oss-120b | SambaNova Free tier with rate limits | 20/min · 20/day | 200K/day | See provider page | no | docs.sambanova.ai checked 2026-10-25 |
| DeepSeek-V3.2 | SambaNova Free tier with rate limits | 20/min · 20/day | 200K/day | Preview model: evaluation only, may be removed at short notice. | no | docs.sambanova.ai checked 2026-10-25 |
| gemma-4-31B-it | SambaNova Free tier with rate limits | 20/min · 20/day | 200K/day | Preview model: evaluation only, may be removed at short notice. | no | docs.sambanova.ai checked 2026-10-25 |
| Free mode (models and limits per your Limits page) free-mode | Mistral Free tier with rate limits | n/p | n/p | See provider page | no | docs.mistral.ai checked 2026-10-25 |
| Mistral Moderation 2 mistral-moderation-2603 | Mistral Free tier with rate limits | n/p | n/p | Listed at $0 per million tokens on the pricing page. | no | docs.mistral.ai checked 2026-10-25 |
| Cohere: North Mini Code (free) cohere/north-mini-code:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 256,000 token context | no | openrouter.ai checked 2026-10-25 |
| Dots Studio: Dots3-Note Preview (free) dots-studio/dots-3-note-preview:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 512,000 token context | no | openrouter.ai checked 2026-10-25 |
| Google: Gemma 4 26B A4B (free) google/gemma-4-26b-a4b-it:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 262,144 token context | no | openrouter.ai checked 2026-10-25 |
| Google: Gemma 4 31B (free) google/gemma-4-31b-it:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 262,144 token context | no | openrouter.ai checked 2026-10-25 |
| LiquidAI: LFM2.5-2.6B (free) liquid/lfm-2.5-2.6b:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 65,536 token context | no | openrouter.ai checked 2026-10-25 |
| NVIDIA: Nemotron 3 Nano Omni (free) nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 256,000 token context | no | openrouter.ai checked 2026-10-25 |
| NVIDIA: Nemotron 3 Super (free) nvidia/nemotron-3-super-120b-a12b:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 262,144 token context | no | openrouter.ai checked 2026-10-25 |
| NVIDIA: Nemotron 3 Ultra (free) nvidia/nemotron-3-ultra-550b-a55b:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 1,000,000 token context | no | openrouter.ai checked 2026-10-25 |
| NVIDIA: Nemotron 3.5 Content Safety (free) nvidia/nemotron-3.5-content-safety:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 128,000 token context | no | openrouter.ai checked 2026-10-25 |
| NVIDIA: Nemotron 3.5 Lightning (free) nvidia/nemotron-3.5-lightning:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 1,000,000 token context | no | openrouter.ai checked 2026-10-25 |
| NVIDIA: Nemotron 3.5 Super VL (free) nvidia/nemotron-3.5-super-vl:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 1,000,000 token context | no | openrouter.ai checked 2026-10-25 |
| Poolside: Laguna S 2.1 (free) poolside/laguna-s-2.1:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 262,144 token context | no | openrouter.ai checked 2026-10-25 |
| Poolside: Laguna XS 2.1 (free) poolside/laguna-xs-2.1:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 262,144 token context | no | openrouter.ai checked 2026-10-25 |
| Recursive: Flint (free) recursive/flint-v1.0:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 262,144 token context | no | openrouter.ai checked 2026-10-25 |
| Thinking Machines: Inkling Small (free) thinkingmachines/inkling-small:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 1,048,576 token context | no | openrouter.ai checked 2026-10-25 |
| Thinking Machines: Inkling (free) thinkingmachines/inkling:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 1,048,576 token context | no | openrouter.ai checked 2026-10-25 |
| XingChen: Xing4.0 29B A4B (free) xingchen-agi/xing4.0-29b-a4b:free | OpenRouter Free models | 20/min · 50/day | n/p | 1,000 req/day after buying 10+ credits 262,144 token context | no | openrouter.ai checked 2026-10-25 |
| Big Pickle big-pickle | OpenCode Free models | n/p | n/p | Stealth model; the team uses the free period to collect feedback. | not stated | opencode.ai checked 2026-10-25 |
| GLM-5.3-Flash Free glm-5.3-flash-free | OpenCode Free models | n/p | n/p | Zero-retention provider, not used for training. | not stated | opencode.ai checked 2026-10-25 |
| DeepSeek V4.1 Flash Free deepseek-v4.1-flash-free | OpenCode Free models | n/p | n/p | Zero-retention provider, not used for training. | not stated | opencode.ai checked 2026-10-25 |
| Beam Free beam-free | OpenCode Free models | n/p | n/p | Not used for training; kept 30 days for safety. Not available in some countries. | not stated | opencode.ai checked 2026-10-25 |
| Flint 40B Free flint-40b-free | OpenCode Free models | n/p | n/p | Used to collect feedback and improve the model. | not stated | opencode.ai checked 2026-10-25 |
| MiMo-V2.6-Flash Free mimo-v2.6-flash-free | OpenCode Free models | n/p | n/p | Used to collect feedback and improve the model. | not stated | opencode.ai checked 2026-10-25 |
| Nemotron 3 Ultra Free nemotron-3-ultra-free | OpenCode Free models | n/p | n/p | Used to collect feedback and improve the model. | not stated | opencode.ai checked 2026-10-25 |
| Nemotron 3.5 Lightning Free nemotron-3.5-lightning-free | OpenCode Free models | n/p | n/p | Used to collect feedback and improve the model. | not stated | opencode.ai checked 2026-10-25 |
| Nemotron 3.5 Super VL Free nemotron-3.5-super-vl-free | OpenCode Free models | n/p | n/p | Used to collect feedback and improve the model. | not stated | opencode.ai checked 2026-10-25 |
| Jev 1.13 Free jev-1.13-free | OpenCode Free models | n/p | n/p | Limited-time free model. | not stated | opencode.ai checked 2026-10-25 |
| GLM-4.7-Flash glm-4.7-flash | Z.ai Free models | n/p | n/p | No limits published | not stated | docs.z.ai checked 2026-10-25 |
| GLM-4.5-Flash glm-4.5-flash | Z.ai Free models | n/p | n/p | No limits published | not stated | docs.z.ai checked 2026-10-25 |
| GLM-4.6V-Flash glm-4.6v-flash | Z.ai Free models | n/p | n/p | Vision model. | not stated | docs.z.ai checked 2026-10-25 |
| Workers AI catalog (except paid-only models) workers-ai | Cloudflare Free credit / quota | n/p | n/p | 10,000 Neurons/day | not stated | developers.cloudflare.com checked 2026-10-25 |
| Free Tier eligible models (subset of the catalog) free-tier-models | Vercel Free credit / quota | n/p | n/p | $5 credit/month | not stated | vercel.com checked 2026-10-25 |
| Generative APIs serverless models generative-apis | Scaleway Free credit / quota | n/p | n/p | 1M tokens + 60 min audio | not stated | www.scaleway.com checked 2026-10-25 |
| Chat models (Command A+, A Reasoning, A Translate, A, R+, R, R7B, North Mini Code) command-models | Cohere Free tier with rate limits | 20/min | n/p | 1,000 calls/month (all endpoints) | not stated | docs.cohere.com checked 2026-10-25 |
| Embed embed | Cohere Free tier with rate limits | n/p | n/p | 2,000 inputs/min; 1,000 calls/month | not stated | docs.cohere.com checked 2026-10-25 |
| Rerank rerank | Cohere Free tier with rate limits | 10/min | n/p | 1,000 calls/month | not stated | docs.cohere.com checked 2026-10-25 |
| Audio transcription transcribe | Cohere Free tier with rate limits | 5/min | n/p | 1,000 calls/month | not stated | docs.cohere.com checked 2026-10-25 |
| NIM API catalog models (prototyping) nim-api-catalog | NVIDIA Free tier with rate limits | n/p | n/p | No limits published | not stated | docs.api.nvidia.com checked 2026-10-25 |
Providers
| Provider | Free offer | Card needed? | Limits in short |
|---|---|---|---|
| Google Gemini API | 9 models / offers | no | Limits are set per model and per project and are not published on the docs page; Google says to view your active limits in Google AI Studio. Requests per day reset at midnight Pacific time. |
| Groq | 10 models / offers | not stated | Limits apply at the organization level. Cached tokens do not count towards limits. |
| Cerebras Inference | 2 models / offers | not stated | Free tier: 5 requests/min, 30K uncached tokens/min, 90K total tokens/min, 1M tokens/hour and 1M tokens/day per model. |
| SambaNova Cloud | 5 models / offers | no | Free tier: 20 requests/min, 20 requests/day and 200K tokens/day per model. |
| Mistral AI (La Plateforme) EU provider | 2 models / offers | no | The numbers are shown on the Limits page in Mistral's console, not in the public docs. |
| OpenRouter | 17 models / offers | no | Free variants: 20 requests/min; 50 requests/day if you bought less than 10 credits in total, 1,000 requests/day with at least 10. A negative balance blocks free models too. |
| OpenCode Zen | 10 models / offers | not stated | No numbers on the public page |
| Z.ai (GLM) | 3 models / offers | not stated | No numbers on the public page |
| Cloudflare Workers AI | 1 model / offer | not stated | 10,000 Neurons/day, reset at 00:00 UTC. How many tokens that buys depends on the model's Neuron rate (for example llama-3.2-1b: 2,457 Neurons per million input tokens). |
| Vercel AI Gateway | 1 model / offer | not stated | $5/month of credit; free-tier requests have lower per-model rate limits than the paid tier (numbers not published on the pricing page). Buying credits moves the team to the paid tier and ends the monthly free credit. |
| Scaleway Generative APIs EU provider | 1 model / offer | not stated | 1M tokens and 60 minutes of audio transcription. The page does not say whether this resets. |
| Cohere | 4 models / offers | not stated | Trial keys: 1,000 API calls/month in total; chat models 20 requests/min, Embed 2,000 inputs/min, Rerank 10 requests/min, audio transcription 5 requests/min. |
| NVIDIA NIM API (build.nvidia.com) | 1 model / offer | not stated | No numbers on the public page |
Gone or not free anymore
| Service | Status | Details |
|---|---|---|
| GitHub Models | Retired | Fully retired on July 30, 2026: playground, catalog, inference API and BYOK are no longer available. Source, checked 2026-10-25. |
| Hugging Face Inference Providers | No free credits for free accounts | Free users get no monthly credits and must buy credits; PRO users and Team/Enterprise seats get $2.00 per month. Source, checked 2026-10-25. |
Good to know
- Google (Gemini) and Mistral show your actual free limits only inside their consoles; the public docs don't list the numbers.
- Google says free-tier content is used to improve its products. Check the data note on each provider page before sending anything private.
- OpenRouter's :free models allow 50 requests a day until you've bought at least 10 credits in total, then 1,000.
- OpenCode Zen's free chat models only work inside the OpenCode client.
Free-tier changes in your inbox
An occasional short email when free tiers, free models or rate limits change, each change linked to its official source. No spam. Every email has a one-click unsubscribe link.