This is a copy of 'Qwen/Qwen-VL-Chat-Int4' but with the possibility to pass the image tensors as a parameter, not downloading them from internet. model.forward(**inputs) -> model.forward(**inputs, image_list)

Downloads last month
3
Safetensors
Model size
4.05B params
Tensor type
F16
·
I32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support