Model Card for Qwen2.5-VL-7B-ViT

This is an unofficial, extracted NaViT Vision Encoder from Qwen2.5-VL-7B.
This model is used for MGM-Omni.

Downloads last month
1,028
Safetensors
Model size
677M params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for wcy1122/Qwen2.5-VL-7B-ViT

Finetuned
(573)
this model

Collection including wcy1122/Qwen2.5-VL-7B-ViT