3 ms·

I would agree, but I want to add that I have real issues with combination of opencode plus slow inference (4-5tok/s). I get weird interruptions. I can only gue

by mhaberl 24d ago

I would agree, but I want to add that I have real issues with combination of opencode plus slow inference (4-5tok/s). I get weird interruptions. I can only guess its related to some kind of timeouts in the harness or something. Its not a problem of the model of course, but it seems impractical atm. I wonder if anyone else had this kind of thing happening.

redox99 24d ago
I think pi handles it better
- fhn 24d ago
  I just tried pi, absolute junk compared to hermes
  - mhaberl 24d ago
    I dont get it, so for slow inference its: hermes > pi > opencode ?? or is hermes better in general? habent tried it hermes/pi yet
ygjb 24d ago
For OSs harnesses this seems like a good thing to point a paid model at fixing if you don't want to dig in yourself?
- mhaberl 24d ago
  I might do that myself in the end checked the bug reports and PRs first though, there are a few related, nothing merged yet I ran opencode task again last night since it's slow and this morning i got a "SSE read timed out" logs show tokens still streaming in steadily (every 3-4s, 46,818 tokens in) right up to the moment opencode disconnected opencode cut a response that was actively generating (not a stalled one)
  - mhaberl 22d ago
    [dead]