Bug Report: Loss of Dynamic Rate Limit Update for GPT-Image Models via API

[OrionCodeDev]

Confirm this is an issue with the Python library and not an underlying OpenAI API

• This is an issue with the Python library

Describe the bug

As of today, we have lost the ability to dynamically update the rate limit for the gpt-image-1 and gpt-image-1-mini models via the API. Previously, we were able to adjust the rate limits for image generation per minute. However, any attempts to modify the rate limit for these models now result in an error indicating that the rate limit for these models is not set for the organization.

Issue Details:

When making a POST request to update the rate limit for the gpt-image-1 model, we received the following response:

{

"error": {

    "message": "The rate limit for model gpt-image-1 is not set for your organization",

    "type": "invalid_request_error",

    "param": null,

    "code": "rate_limit_does_not_exist_for_org_and_model"

}

}

This error implies that the rate limit configuration for the gpt-image-1 model is not available for our organization, even though this functionality was previously working without issue.

Steps to Reproduce:

1. Attempt to update the rate limit for the gpt-image-1 model using the following POST request:

POST v1/organization/projects/proj_XXXX/rate_limits/rl-gpt-image-1

Request Body:

{

   "max_images_per_1_minute": 500

}

2. The response returns an error:

{

   "error": {

       "message": "The rate limit for model gpt-image-1 is not set for your organization",

       "type": "invalid_request_error",

       "param": null,

       "code": "rate_limit_does_not_exist_for_org_and_model"

   }

}

Expected Behavior:

The API should allow us to dynamically update the rate limit for the gpt-image-1 and gpt-image-1-mini models, including the ability to modify the max_images_per_1_minute field.

Actual Behavior:

The API returns an error indicating that the rate limit configuration for the gpt-image-1 model is not available for our organization, which prevents us from adjusting the image generation rate limits.

Impact:

This issue directly impacts our ability to manage and scale our usage of the GPT-Image models, as the dynamic rate limit updates for image generation are no longer functioning as expected.

Environment:

• Affected models: gpt-image-1, gpt-image-1-mini

Logs/Errors:

{

"error": {

    "message": "The rate limit for model gpt-image-1 is not set for your organization",

    "type": "invalid_request_error",

    "param": null,

    "code": "rate_limit_does_not_exist_for_org_and_model"

}

}

Additional Information:

• The issue seems to have appeared today and might indicate a change in the availability of rate limits for these models within the organization settings.

• We previously had the ability to update rate limits for these models dynamically, but this functionality appears to be broken now.

Request:

We kindly ask for assistance in resolving this issue by restoring the ability to dynamically update the rate limits for the gpt-image-1 and gpt-image-1-mini models via the API, or to clarify if there has been a change in the API functionality for these models.

To Reproduce

^^

Code snippets

OS

Linux

Python version

latest

Library version

latest