Prince Canuma comments

Results 572 comments of


                                            Prince Canuma

Model Performance & Compatibility Report (2025-11-14)

This is really great! Could you include the image and prompts as well ?

Model Performance & Compatibility Report (2025-11-14)

I can't find the images

Model Performance & Compatibility Report (2025-11-14)

Thanks, this last report is nicely detailed Let me breakdown into issues

Model Performance & Compatibility Report (2025-11-14)

@jrp2014 Could you open an issue for the Qwen2-VL OOM with inputs and outputs and entire traceback?

Command A Vision responses contain only "<PAD>" tokens

Thanks! Yes, command-a still has a bug. I haven't been able to figure it out yet but I will work out. On paper the implementation is identical to the aya...

Add sequence number handling to streaming events in OpenAI endpoint

Thanks @dwohlfahrt! Could you share a bit more about this sequence number? And why is it important?

Mapping VLA's to port to MLX by the king!

Thanks, this helps a lot!

Fixed import of RepositoryNotFoundError

Thank you very much for the PR @cmcmaster1! However, this fix throws an error for me on hugginface-hub v0.24.2 but it works fine on v0.25.1. Could pin it's version on...

Fixed import of RepositoryNotFoundError

After that is done, it will be ready to merge.

Fixed import of RepositoryNotFoundError

Hey 👋