Nicholas King

Very cool, how did you run that? Is it part of Github Actions?

Here's the gist of the logic flow of how data is sent to the server, and how it's received and put in the Redux store. 

+1 to this request. There are devs I work with who don't use VSCode as their editor but would still find devcontainers useful in their workflow. An interactive shell is...

Thanks for your reviews Hudson & Harsha! I am picking this PR back up now and will make revisions based on your feedback. I'm also going to work on integrating...

Closing this PR since Hudson has implemented an alternative system - it can remain a reference for local models like Phi 3 vision in the future though.

After doing some more debugging I found that I can repro this much more often when using a higher temperature: ` lm += gen("test1", max_tokens=100, temperature=0.8)` This can repro the...

The bug also seems to repro more often with a longer prompt and higher `max_tokens`

Hi @microdev1, this is a great start on Ollama support in Guidance. Thanks for your contribution! I tested the Ollama class and was able to run a model using it,...

This is very helpful. I would suggest adding some introductory prose about what the notebook covers (harmony format examples, tool use, lark grammar) as well as requirements (active GPT-OSS deployment)....

Thank you for the feedback. We are actively working on a full-stack rework of multimodal support in Guidance. Part of this rework involves reformatting the way that prompt data is...