Prince Canuma

medium.com/@prince.canuma [email protected]

kulissiwa.com Poland MLOps | LLMs | VLMs | Audio LMs | Ex - ML Research Engineer @arcee-ai

Results 572 comments of


                                            Prince Canuma

Error loading model: Module.load_weights() got an unexpected keyword argument 'weights'

What version on MLX-audio are you using ?

Error loading model: Module.load_weights() got an unexpected keyword argument 'weights'

Closing it as stale

Training?

Not yet. Still need to add a few models, add batch processing and create a standard trainer

could run it without network?

Yes, it can run without internet. Unless you are downloading new voices. Could you share a reproducible example?

could run it without network?

Thanks! I see the issue. I should check local cache before hitting HF for voices. It will be fixed in the next release :)

LLaMA 3.2 11B Vision Support

It's coming

LLaMA 3.2 11B Vision Support

What could you elaborate and show some examples?

LLaMA 3.2 11B Vision Support

Not yet. My plate if full with #41

LLaMA 3.2 11B Vision Support

I'm working on porting it to MLX.

when can support Qwen3-Omni ? this is realtime video and vocie chat model. very useful

Thanks @zujian-y! We can support it and it has been on the backlog the longest

‹
1
2
...
44
45
46
47
48
49
50
...
57
58
›