Greg Tanaka

Results 8 comments of Greg Tanaka

I have a M3 pro with 36GB of memory. I can run the mixtral:8x7b-instruct-v0.1-q3_K_L (20GB) with the GPU and there is 10GB of free memory when it runs, but if...

I tried pip install --pre xformers==0.0.17.dev461 It about the same runtime BTW, I am also using deepspeed 0. It is about the same runtime without deepspeed.

Here is my nvidia-smi: Mon Feb 20 08:07:29 2023 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 525.60.13 Driver Version: 525.60.13 CUDA Version: 12.0 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr....

Thanks for your work. I just tried this. Unfortunately, it didn't work for me on my m1 max with 32GB. Here is what I did: pip install git+https://github.com/openai/whisper.git@refs/pull/382/head No errors...

I am just getting noise: ![prediction_0](https://user-images.githubusercontent.com/24632716/202891521-332506e9-4b9c-41f8-a7c2-51513a7de042.png) I think something is wrong with the h5 files

It would be great if the latest openai libraries could be supported. Most of the other libraries like Llama Index and Langchain support the latest