Image analysis does not work when using o1-mini or o3-mini with agents.

[distardao]

What happened?

When using o1-mini or o3-mini directly for image understanding tasks, everything worked fine. However, when I switched to using these APIs with agents, an error occurred:
"com_error_invalid_request_error"

Version Information

version 0.7.7

Steps to Reproduce

1. Create an agent with o3-mini
2. Feed an image to analyze

What browsers are you seeing the problem on?

No response

Relevant log output

Screenshots

No response

Code of Conduct

• I agree to follow this project's Code of Conduct

[danny-avila]

Regular endpoint switches the model to use a vision model if the currently selected one doesn't have this capability. When using Agents, it does not switch the model from what you configured.

https://platform.openai.com/docs/models/o1-mini
https://platform.openai.com/docs/models/o3-mini

Only o1 (not mini) and o1-pro support image inputs