Ar770 comments

Repositories
Issues
Comments

Results 4 comments of


                                            Ar770

I want to use whisper-at models

Possible use faster-whisper (https://github.com/guillaumekln/faster-whisper) backend?

@YuanGongND, I want to make a PR with a faster-whisper implementation. Could you please highlight what needs to be done to implement it?

Only one line in the output srt file!

Same issue with me.

[Bug] Exception: cublasLt ran into an error! during fine-tuning LLM in 8bit mode

Trying to run today on a H100 instance, confirmed installation of 0.40.1 that I saw that was supposed to work now with this GPU, I still get: ``` --------------------------------------------------------------------------- Exception...