ModelQwen3-VL-32B
Qwen3-VL-32B
A 32-billion-parameter multimodal vision-language model released by Alibaba's Qwen team, supporting image understanding and visual reasoning.
Timeline (last 90 days)
Oct 17
On Multimodal RewardBench 2, JEV-27B-VL averages 63.8 across four tasks, close to GPT-4.1 and Qwen3-VL-32B but behind GPT-5 and Gemini 3 Pro; its image editing score of 57.8 is a clear weak spot.
Unverified50%