VOOZH about

URL: https://huggingface.co/mistralai/Pixtral-Large-Instruct-2411/discussions/3

⇱ mistralai/Pixtral-Large-Instruct-2411 · Upload transformers version

Upload transformers version

#3
by ArthurZ HF Staff - opened
No description provided.
ArthurZ changed pull request status to open
pandora-s changed pull request status to merged

Converted model does not work for now let me check!

Bias were missing for the multimodal projection layers

They seem to still be missing?

👁 image.png

@pandora-s
When you updated mistralai/Mistral-Nemo-Instruct-2407 2 weeks ago, it looks like you broke it. Can you please see about fixing it for everyone? Thanks

deleted

@pandora-s
When you updated mistralai/Mistral-Nemo-Instruct-2407 2 weeks ago, it looks like you broke it. Can you please see about fixing it for everyone? Thanks

Geesh, you going to spam all of HF with this?

Any updates @ArthurZ ?

The safetensors are still not working properly.

I Quantised it to 4bit in MLX and it doesn’t spit out proper English

Converted model does not work for now let me check!

@ArthurZ Any updates on this?

new PR (upload in progress): #23

This comment has been hidden (marked as Off-Topic)
This comment has been hidden (marked as Off-Topic)

· Sign up or log in to comment