All fixes · OpenAI / OpenAI-compatible · 404 · Model limits

“The model does not exist or you do not have access to it” (404): How to Fix

The model `…` does not exist or you do not have access to it

The API could not match the model id to a model your key may use: the name is misspelled or retired, your account has no access to it yet, or the base URL points at a different provider than the one that serves it. Check the exact model id, your access, and that you’re calling the right endpoint and base URL.

What causes it

  • Typo or outdated model name (models get renamed and retired).
  • Your key/org doesn’t have access to that model yet.
  • Wrong base URL: calling one provider’s endpoint with another’s model id.

How to fix it

  1. Verify the exact model id against the provider’s current model list.
  2. Confirm your key/org has access to that model.
  3. Check the base URL matches the provider that serves the model.

Keep working on a fixed monthly plan

On Standard Compute the model is always just “standardcompute”, a single id that routes each request to a frontier or an efficient model from the catalog. You never chase renamed or deprecated model names. Set the base URL to https://api.stdcmpt.com/v1 and you’re done.

OpenAI-compatible settings
Base URL  https://api.stdcmpt.com/v1
API key   YOUR_API_KEY
Model     standardcompute

Plans from $19 a month with $19 of compute. Eligible new accounts can try $0.25 of compute first, no card. Trial terms

FAQ

Why did a model name that used to work stop working?

Providers deprecate and rename models regularly. A name that worked last month can 404 today. Pinning to a single routed identifier avoids the churn.

Related errors