About free-model.com

free-model.com is a free directory of free LLM APIs. We catalog models from Google AI Studio, Groq, NVIDIA NIM, OpenRouter, Mistral, Cohere, Cerebras, Cloudflare Workers AI, and more — so developers can find the best free AI models without digging through GitHub READMEs.

Why We Built This

There are hundreds of free LLM APIs available, but no single place to compare them side by side. Developers waste hours reading docs and testing endpoints just to find a free coding model. free-model.com puts all the information — context windows, rate limits, modality, API compatibility — in one searchable, filterable table.

Contribute

This site's model data is open source — synced and updated daily on . Found a missing model, incorrect data, or have feedback? Open an issue or submit a PR — stars welcome! You can also reach out on .

How It Works

Every day, our automated scripts pull the latest model data from provider APIs (OpenRouter, NVIDIA NIM, Google AI Studio, Groq, Mistral, Cohere, Cerebras, Cloudflare Workers AI, and more). We check which models are online, which have gone offline, and whether rate limits or free status have changed. The site rebuilds automatically with fresh data.

How We Verify Models — Manually Tested, Not Just Copied

Most free LLM API directories copy model lists from GitHub READMEs without testing whether the APIs actually work. We do things differently — every model from OpenRouter and NVIDIA NIM is manually tested via live API calls before being listed as "online."

Our verification process:

  • Live API testing — We send real chat completion requests to each model endpoint using a valid API key. If the model returns a successful response, it's marked as verified and online.
  • Error diagnosis — When a model fails (404, 403, timeout, credit required), we investigate the root cause. Some models have incorrect IDs in their source data (e.g., NVIDIA NIM uses slash format deepseek-ai/deepseek-v3 but READMEs list dashes). We fix these at the data source.
  • Status accuracy — Models that consistently fail are marked as offline or unavailable. No "probably works" guesswork.
  • Daily automated refresh — Our sync scripts pull fresh data from provider APIs every day, checking for new models, status changes, and updated rate limits.

The result: when you see a green "Verified" badge on a model, it means someone actually sent an API request to that model and got a working response — not just that some README said it was free.

Our Principles

  • No fictional data. Every model on free-model.com comes from a real provider API or a verified open-source dataset. We don't fabricate models, specs, or rate limits.
  • No credit card gate. We highlight which providers don't require a credit card — because "free" with card verification isn't truly frictionless.
  • Direct links, no reselling. We link you to the provider's own signup page. We don't resell API keys, shared accounts, or proxy access.
  • Transparent about data freshness. Every page shows when data was last refreshed. Model statuses can change daily — we flag stale data openly.

Data Sources

Our model data is aggregated from:

Model data is updated daily via automated GitHub Actions. Provider status and rate limits may change — always check the provider's official page for the latest information.