I would like to benefit from the 2 million token window (or larger) that some of the XAI models provide through their own infrastructure and APIs. As of now when i add my own KEY and use the Grok models from XAI . I am limited to a much smaller window size (120k). I am not clear why this limitation since these requests do not pass through the Microsoft or Github.com infrastructure.

I hope this can be added to support use cases where larger context window is needed.