I am trying to run a script that uses Batch API for gpt-4o-2024-08-06 and recently it started to fail with the error:
{“id”: “batch_req_67f76b26f2b481908203aec37419c330”, “custom_id”: “9”, “response”: {“status_code”: 403, “request_id”: “8f8cb5d99d63a92862be4130a1ee6349”, “body”: {“error”: {“message”: “Project proj_pANHoRc5qch2oOrsH16jZ20j does not have access to model gpt-4o-2024-08-06-batch”, “type”: “invalid_request_error”, “param”: null, “code”: “model_not_found”}}}, “error”: null}
Has anyone had a similar issue?
Yes, the same has been seen before. There’s some internal model that your request is converted to, ending in -batch, maybe to give the discount.
First thing: Generate and use a new API key to submit the request. That key might then be repopulated with all the necessary internals.
Yes, the same has been seen before. There’s some internal model that your request is converted to, ending in
-batch, maybe to give the discount.
First thing: Generate and use a new API key to submit the request. That key might then be repopulated with all the necessary internals.
We had to re-enable limits for the models as well.
Project (Default Project) -> Limits page and re-enable the specific model.
@_j Thank you for your help!