Qwen2.5 VL! Qwen2.5 VL! Qwen2.5 VL!
Alibaba released Qwen2.5-VL with 3B, 7B, and 72B variants, offering major vision-language improvements over Qwen2-VL with open-source weights.
Excerpt
QWEN CHAT GITHUB HUGGING FACE MODELSCOPE DISCORD
We release Qwen2.5-VL, the new flagship vision-language model of Qwen and also a significant leap from the previous Qwen2-VL. To try the latest model, feel free to visit Qwen Chat and choose Qwen2.5-VL-72B-Instruct. Also, we open both base and instruct models in 3 sizes, including 3B, 7B, and 72B, in both Hugging Face and ModelScope.
The key features include:
Understand things visually: Qwen2.
Read at source: https://qwenlm.github.io/blog/qwen2.5-vl/