Exactly what I was thinking - even on frontier or near-frontier models I still see my agents get stuck in these pointless loops where it's very obvious to me what they need to do to get "unstuck".
For our local Qwen, your setup works great out of the box!
Did you notice any particular guardrails firing? Always curious about things I haven't tested on - especially if it has a different shape.
It should work with opencode using the proxy server or middleware method right? Any tips?
Does this need a GPU to work? Or is it CPU only? I ask because I plan to try to run this using Docker. But I have a modest RTX 5070 12GB VRAM.
Or maybe I could use opencode as a remote backend too?
I'm thinking of trying the OpenAI-compatible provider route: https://opencode.ai/docs/providers/#custom-provider