gottlike

Recent Crystal compiler (v0.31.1) throws deprecation warning and suggests ".decode" or ".decode_www_form"

WizardCoder improvement request: Train on OpenAPI specs

1

WizardCoder currently has weaknesses in understanding and using [OpenAPI specs](https://swagger.io/specification/) in order to create code that queries the specified API to answer a specific question. It would be great to...

Thank you for providing two excellent models to the open source community. Since you created WizardCoder after WizardLM I wonder: Do you have any plans on optimizing those models further...

Whisper support

31

Is support for Whisper on the roadmap? Something like https://github.com/ggerganov/whisper.cpp would be great.

Paper: New algorithm for pruning

2

Just a FYI of a new discovery that should be super relevant for you guys: https://twitter.com/aaquib_syed1/status/1714386165237776653

I'm POSTing a JSON array of objects that can have all sorts of characters. I encountered the following error: ``` ``` The error message stops at the point, where the...

Support for Command-R model from Cohere

4

Since it's a pretty great model, I would love being able to train it: https://huggingface.co/CohereForAI/c4ai-command-r-v01

Cohere's Command-R model support

1

Would be great to be able to serve Command-R with DeepSpeed: https://huggingface.co/CohereForAI/c4ai-command-r-v01

Hi, I'm using your wrapper to access the Adobe Survey API which sometimes takes a long time (minutes) to return data. In this case I need to specify a timeout...

Consider methods to prevent hallucinations and incorrect answers

5

Just wanted to put this out here for your consideration: https://www.youtube.com/watch?v=LO3U5iqnTpk A great and advanced summary of things that can go wrong with LLM search and ways of improving/safeguarding results....