Please explain the motivation behind the feature request.

This new model is cost effective and fast, especially good at writing code.

Describe the solution you'd like

Please officially support it, especially its 256K context window, not 131K.

Describe alternatives you've considered

With #4119 workarounds the unusable issue, grok-code-fast can be specified as a "custom" model, but this way only 131K context is used, not its actual 256K

Additional context

Please incorporate #4119 to solve #4177 too.

• I have verified this does not duplicate an existing feature request