Richard Guo

Hey @Chae4ek! The parameters are needed to set the prompt context correctly. They typically will not require any changes on the user side and as far as I know work...

Chiming in here on auto-suggest / installation path. We have a C library method for deciding correct model to load based on magic numbers for the model binaries. We haven't...

I think having filename be part of model_path is reasonable and no strong feelings there if we want to see that change in the `gpt4all` package. I do think keeping...

PR looks good to me @dev2049!

Doesn't look like I have permission to resolve conversations here? PR lgtm!

The model loading statistics and Found model file/downloading messages all happen during model instantiation (`gpt4all.GPT4All(model_name)`). The `verbose=False` option helps control outputs during chat_completion and will not impact this. I can...

@jpzhangvincent there's a large PR consolidating ggml versions currently in the process of being merged into main. This will be going in after that :)

@manyoso @niansa ready for another review

@apage43 Fixed - how does it look now? Edit: prefacing this with I'm not a C++ developer :'). So let me know if there are better practices / concerns I...

Hmm just ran conversion script again and is working fine for me. Which OS are you working on?