I’m using Qwen3.6-27B running on my server with llama-server for AI coding with OpenCode. Sometimes for some reason, the response stops when its reasoning like if it has finished outputting the full response. I have to type “continue” and it continues working like if nothing happened. It doesn’t sho
