freellmpool › reference
These 24 providers offer a free LLM API tier you can call today. Three need no API key at all, and LLM7 is key-optional. The table shows whether a key is required, roughly how many models the free tier exposes, and a free-tier signal where known. The catch with using several is that each has its own SDK and limits — the open-source freellmpool pools all of them behind one OpenAI-compatible endpoint with failover, so you get their combined free capacity through one interface.
| Provider | API key? | Free models* | Free-tier signal** |
|---|---|---|---|
| Pollinations | No (keyless) | 2 | Anonymous; rate-limited per IP |
| OVHcloud AI Endpoints | No (keyless) | 17 | Anonymous tier (chat + embeddings + Whisper) |
| LLM7 | Optional | 3 | ~60/hour without a token |
| Kilo Gateway | No (keyless) | 11 | Anonymous; ~200 req/hour per IP |
| OpenCode Zen | No (keyless) | 0 | Cataloged routes disabled by default pending opt-in |
| Groq | Yes (free) | 12 | ~1,000 req/day (per model) |
| Cerebras | Yes (free) | 3 | ~14,400 req/day (large cap) |
| Google Gemini | Yes (free) | 3 | ~1,500 req/day |
| OpenRouter | Yes (free) | 20 | ~1,000 req/day on :free models |
| GitHub Models | Yes (any PAT) | 22 | ~150 req/day |
| NVIDIA NIM | Yes (free) | 39 | Free credits / rate-limited |
| Cloudflare Workers AI | Yes (free) | 24 | Daily Neuron allowance |
| Mistral | Yes (free) | 43 | Free experimentation tier |
| Cohere | Yes (free) | 21 | Trial keys, rate-limited |
| SambaNova | Yes (free) | 0 | Free tier, rate-limited |
| Z.ai / Zhipu GLM | Yes (free) | 2 | ~1,000 req/day |
| Ollama Cloud | Yes (free) | 8 | Qualitative free limits |
| LongCat (Meituan) | Yes (free) | 0 | Preview route retired; no active route |
| Hugging Face (router) | Yes (free) | 22 | Monthly included Inference credits |
| Aion Labs | Yes (free) | 5 | 20K tokens/day; no card |
| ModelScope API Inference | Yes (free) | 7 | 2,000 calls/day total; dynamic per-model caps |
| Morph | Yes (free) | 3 | 200 requests/month |
| Vercel AI Gateway | Yes (free) | 5 | $5 recurring monthly Hobby credit |
| SiliconFlow | Yes (free) | 1 | Fixed free-model limits; identity verification required |
* Enabled chat, embedding, and transcription routes in freellmpool's catalog; varies as tiers change. ** Free-tier signals are approximate and change without notice — always check the provider's current docs. freellmpool tracks your real per-day usage as you go.
pip install freellmpool
freellmpool ask "Explain the CAP theorem in one sentence." # keyless, picks a free provider
freellmpool providers # see which tiers you've configuredAdd each provider's free key as an env var to grow capacity; signup links are in the accounts guide. See also: using multiple free LLM APIs together · best free LLM API gateway.
As of 2026, free API tiers are offered by Groq, Cerebras, Google Gemini, NVIDIA, OpenRouter, GitHub Models, Cloudflare, Mistral, Cohere, SambaNova, Z.ai/Zhipu, Ollama Cloud, LongCat, LLM7, Pollinations, OVHcloud, Kilo Gateway, OpenCode Zen, Hugging Face, Aion Labs, ModelScope, and SiliconFlow. Pollinations, OVHcloud, and Kilo Gateway work with no API key by default; OpenCode Zen is cataloged keyless but disabled by default.
Among the common free tiers, Cerebras advertises one of the largest daily request caps, and providers like Groq and OpenRouter offer roughly a thousand requests/day on free models. Limits change often — pool several with freellmpool so a single cap isn't a bottleneck.