← All models

Qwen3.5-Flash (2026-02-23)

Alibaba · 2026-02-23

The February 23 hosted Qwen3.5-Flash snapshot is Alibaba's efficient vision-language model with hybrid linear attention and sparse mixture-of-experts architecture. It accepts text, images, and video, returns text, and supports optional thinking with a 1,000,000-token context. Alibaba's lifecycle documentation records API availability on February 23, separately from the following day's medium-model family announcement.

Benchmark scores