kyutai / CASA-Qwen2_5-VL-3B
Vision4.1B3.4K3Updated Sep 6, 2026
Downloads
3.4K
Likes
3
Last updated
Sep 6, 2026
License
cc-by-nc-sa-4.0
This repository contains the model weights for CASA-Qwen25-VL-3B, introduced in the paper CASA: Cross-Attention over Self-Attention for Efficient Vision-Language Fusion.
Tasks & library
- image text to text
- transformers
Tags
- transformers
- qwen2_5vl_ca
- feature-extraction
- image-text-to-text
- conversational
- custom_code
Technical specifications
Parameters4.1B
Librarytransformers
Languagesen
Base modelQwen/Qwen2.5-VL-3B-Instruct
Licensecc-by-nc-sa-4.0