Open IgnisDa opened 3 weeks ago
Hi, @IgnisDa! Thank you for stopping by. And I appreciate the compliment.
I'm thinking of just adding retries in the next releases, but --verbose
is a very nice addition, I'll need to think it through.
LMK when you make a release. I would like to test it.
It seems I get this error often in large files
The repo I am using: https://github.com/ignisda/ryot (The screenshot in the original issue is from a private repo).
Great pointer, thanks for investigating it. It is very probable that this issue is due to the chat history being too large: for now, the conversation does not get rolled up. I'll update you here once I do the rollups and the retries.
@xhjkl LMK if you need help testing.
Hi! Really, really impressive project. I love it!
I often hit this error:
It is non-deterministic (ofc) but I am happy to provide any more details you might need to debug this error.
I also suggest adding a
--verbose
flag so that the exact OpenAI response is returned and we can see what went wrong when runningwell
.