Free tier limits

Free AI API limits by provider

Provider pages built from the 54-provider tracker. Each page lists the published models, rate limits, context windows and card requirements, with source links and a checked date.

How to read these pages

Each provider page is a view of the same 54-provider tracker. The table records what the provider publishes for each model: requests per minute, requests per day, tokens per minute, tokens per day, context window and whether a card is required. A cell marked “console-only” means the provider does not publish the number outside its dashboard; “[VERIFY]” means the tracker has not confirmed it yet. The checked date is shown on every page.

  • Free tier means a recurring allowance that resets daily or monthly.
  • Trial credit means a one-time balance that does not renew.
  • Cloud credit means a signup credit tied to a cloud account and an expiry date.
  • No free API means the provider is included for comparison but has no free allowance.

Why the limits are not comparable

One provider caps requests per day, another caps tokens per day, and a third caps “neurons” or a monthly dollar credit. A 1,000-request allowance can be more generous than a 200,000-token allowance for short chat, and less generous for long documents. Convert the workload to tokens before comparing, then check which cap binds first. The Free API Runway Calculator does that conversion and shows the binding cap.

Providers

More provider pages are added as their limits are verified. If a provider is missing, use the full tracker or the runway calculator.