### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### What happened?

Using latest stable release, `v1.86.2` I am trying to add a model with Mode of Image Generation and Provider of OpenAI but it insists no provider has been provided, failing with this error:

``` LLM Provider NOT provided. Pass in the LLM provider you are trying to call. You passed model=Qwen-Image Pass model as E.g. For 'Huggingface' inference endpoints pass in `completion(model='huggingface/starcoder',..)` Learn more: https://docs.litellm.ai/docs/providers ```

### Steps to Reproduce

Add a db model via the interface with these options: * Provider: OpenAI * LiteLLM Model Name: Custom -> Qwen-Image (testing with Regolo.ai) * Mode: Image Generation * enter credentials * click Test Connect

You will see the above error message.

If you set Mode to Chat then test succeeds, but of course you can't use the model for image generation.

This is the model I am trying to use, but I tried others with other providers, always the same result: * https://regolo.ai/models/qwen-image-2512/

### Relevant log output

```shell 11:45:29 - LiteLLM:DEBUG: utils.py:2…