Free AI API in 2026: about 200 models on one key
Chat, image, embedding and speech models you can call for free through one OpenAI compatible API, what the limits are, and how to start in two minutes.
UnoRouter is a free AI API with about 200 free models (chat, image, embedding and speech) behind one OpenAI compatible key at https://api.unorouter.com/v1. Any model ending in :free costs nothing, with about one request per minute per model.
Most free AI APIs give you one provider's small model and a tight daily cap. UnoRouter puts every free tier we can find behind one OpenAI compatible key: at the time of writing about 200 free models, from GPT, Gemini, Claude Haiku, DeepSeek, GLM and Qwen chat models to image, embedding and speech models.
What you get for free
Any model id ending in :free costs nothing. Today that is more than 100 chat models, about 30 image models, embedding models for search and RAG, and speech models. Free capacity comes from many providers, so the list changes; the models page always shows what is free right now.
Start in two minutes
Create an account with Google, Discord or GitHub, copy your key from the Tokens page, and set the base URL to https://api.unorouter.com/v1. Any OpenAI SDK or compatible client works unchanged: send POST /v1/chat/completions with a model such as gemini-3.8-flash:free. No credit card is needed.
Free image generation API
Image models use the same key on POST /v1/images/generations. flux-2-dev:free is a good start, and there are free anime and photo style models too. You can also try them without code in the built in image studio.
The limits, honestly
Free models allow about one request per minute per model and account; faster calls get HTTP 429 with a Retry-After header. Each free model has its own allowance, so rotating between models is allowed. Some free providers keep or train on prompts; for those you are asked once for consent before any request reaches them. The docs list every limit and error code.
When to pay
Pay when you need steady throughput or a frontier model. The same key then reaches paid models per token, from a $1 top up that never expires, so nothing in your code changes. See pricing.
Get your free key or browse the free models.
Frequently asked questions
Is there a free AI API?
Yes. UnoRouter serves about 200 models for free through one OpenAI compatible key. Any model id ending in :free costs nothing; sign up with Google, Discord or GitHub, no card needed.
Is there a free AI image generation API?
Yes. Free image models such as flux-2-dev:free run on POST /v1/images/generations with the same UnoRouter key.
What are the limits of the free AI API?
About one request per minute per model and account. Faster calls get HTTP 429 with a Retry-After header; wait or switch to another free model.
Trending models this week
What a proxy does in Janitor AI, what makes one good for roleplay, and how the free and paid options compare, including where UnoRouter fits and where it does not.
DeepSeek's own API is paid. Here is how to call DeepSeek models for free through an OpenAI compatible key, what the limits are, and how to use it in your code, Janitor AI or SillyTavern.
Wearing the UnoRouter Discord tag shortens your free-model rate limit window by 25 percent, so a model you could call once a minute you can call every 45 seconds.
Comments
Loading comments