Robert Brennan
Robert Brennan
See https://github.com/OpenDevin/OpenDevin/blob/main/docs/documentation/LOCAL_LLM_GUIDE.md for docs on how to use llama/ollama
The problem here is that the underlying runtime dies (e.g. due to running out of memory) which leaves the HTTP client in the lurch. The HTTP request times out, and...
+1 love the idea, but we should have a script to keep it updated
@powerzbt how did you generate this?
ahh you wrote it by hand? 👍
There are a handful of utilities out there that automate this, e.g. https://github.com/mgechev/github-contributors-list I think we just need to pick one and integrate it. Even if someone needs to run...
@powerzbt I'm going to close this, but if you find a way to automate it let us know!
@qiangli I'm unclear about how this is different from the EventStreamRuntime, which runs inside of docker, using the same server. Can you explain what the goal is here, and why...
Ahh what if we simply add these options to the existing EventStreamRuntime? that will be much easier to maintain
I feel like we should try for `pip install pipenv` before falling back to homebrew. But maybe it doesn't matter?