Unable to use llama offline

[baconsalad]

What happened?

I used to be able to get this to work each update by making some changes but now I cannot get it work at all. It just keeps spitting out Model CODE_QWEN_2_5_1_5B_Q8_0 does not support openai-endpoint-responses: Must be supported to use OpenAI responses params. . The code qwen 2.5 is close enough to qwen3.5 I use now, I don't know why it wants openai anything.

Relevant log output or stack trace

Steps to reproduce

No response

CodeGPT version

0.0

Operating System

Linux

[baconsalad]

I rolled back to 3.7.4 from 3.7.5 and it works again.

[DantSu]

I rolled back to 3.7.4 from 3.7.5 and it works again.

Thanks, I roll back and it's works again ! Many days I thinking that come from Qwen3.5 + Llama.cpp !