YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Qwen 3.5 27B with the vision tower kept in bf16. Inspired by https://huggingface.co/techwithsergiu/Qwen3.5-9B-bnb-4bit

Q: "Does it really matter?"

A: Yes, if you need vision. Here is an unsloth vision lora training comparison:

Downloads last month
161
Safetensors
Model size
28B params
Tensor type
F32
BF16
U8
Inference Providers NEW
This model isn't deployed by any Inference Provider. 馃檵 Ask for provider support