There are two new Dense model sizes added to the Qwen3-VL family.

On October 22nd, Aliyun Tongyi Qianwen announced that the Qwen3-VL family has expanded again, adding two new dense model sizes: 2B and 32B. So far, Qwen3-VL has released a total of four Dense models (2B, 4B, 8B, 32B) and two MoE models (30B-A3B, 235B-A22B). Each model is available in Instruct and Thinking versions, as well as a FP8 quantization version for each of the 12 models. A total of 24 Qwen3-VL open source weight models can be freely downloaded and used for commercial purposes on the Model Hub and Hugging Face platforms.