Possibility to change the timeout value.

[id1740]

With qwen2.5vl everything was fine, but with qwen3vl it very often times out.

[kalebbroo]

How long does it take to generate? I think I have it set to a decent value.

Maybe I need to look at making it adjustable by the user.

[id1740]

How long does it take to generate?

When the model is unloaded, it takes about 30 seconds, otherwise 15-20 seconds.

[id1740]

Just realized that qwen3-vl is a thinking model, that's why it's so slow sometimes.

[kalebbroo]

We have not added support for thinking models yet.

[jtreminio]

#56

[jtreminio]

@kalebbroo good to close this now