Hi there,
Thanks for the awesome work.
I wonder if it is possible to integrate locally deployed large models such as Qwen into the codex harness or Claude code harness. When I try to connect to codex, it doesn't seem to parse <tool_call> from the model output. Is this normal? And is there any ideas to deal with it?
Thanks again.
Hi there,
Thanks for the awesome work.
I wonder if it is possible to integrate locally deployed large models such as Qwen into the codex harness or Claude code harness. When I try to connect to codex, it doesn't seem to parse <tool_call> from the model output. Is this normal? And is there any ideas to deal with it?
Thanks again.