Error with models not supporting prompt cache

[dewrama]

What version of Kimi Code is running?

0.29.1

Which open platform/subscription were you using?

nvidia nim

Which model were you using?

nemotron, minimax

What platform is your computer?

windows, wsl

What issue are you seeing?

This was working fine until the latest update today. I believe Nvidia NIM doesn't support this but I have been using their models for weeks with kimi code. This week's update broke it somehow. Is there a setting I can use to disable it? Thanks.

Model API error
400 Validation: Unsupported parameter(s): prompt_cache_key

What steps can reproduce the bug?

use 3rd party models without prompt_cache support will error

What is the expected behavior?

No response

Additional information

No response

[dewrama]

I think there is still something broken somewhere. I upgraded to newest version and while the prompt_cache error is gone, I am still getting 400 errors for all models on Nvidia NIM except gpt-oss-120b which is OpenAI and compatible. I confimed that the other models work with other harness like opencode.

Anyone else having issues with new kimi on Nvidia NIM?

[wpauli3000]

I have the same problem with nearly all Nvidia NIM models and it still persists with v 0.30:
"Error: [provider.api_error] 400 Validation: Unsupported parameter(s): prompt_cache_key"
Until v 0.27 the error is not showing up.
Please fix it!