Custom Models / Tokens for Custom Endpoints

[Tydorius]

Describe the feature

Features Requested:

Add toggles for overriding the model and token limits via text boxes.

Reasoning:

I was trying to use this with a site that has Deepseek R1 behind an OpenAI compatible API. Some front ends I've used will query the get /models to update their list, and others allow you to type in a custom model name. Anse retains the default list of OpenAI models, meaning that its use is limited.

Paired with this is the maximum token slider with no override options, meaning that even if a custom model could be selected, it could not be used to its full capabilities.

Additional context

No response

Participation

• I am willing to submit a pull request for this feature.