Qwen3.5

The 397B native multimodal agent with 17B active params

Qwen 3.5 is an open-weight native vision-language model built for long-horizon agentic workflows. Its hybrid architecture (linear attention + Mixture-of-Experts) delivers 397B-scale intelligence with the inference speed of a 17B model. Designed for advanced autonomous agents, it combines deep multimodal reasoning, massive context windows, and high efficiency to execute complex real-world tasks with precision.