通义听悟
Great for meeting-heavy teams, students, journalists, and researchers turning speech into knowledge; not for formal archival use without human proofreading. Verification level not recorded ·
Qwen3.5 397B A17B
Rank #92
qwen
qwen/qwen3.5-397b-a17b-20260216
Usage data date 2026-07-27 · updated daily
Usage and specs
Daily rank#92
Daily tokens34.1B
Daily requests4.8M
Daily change
+17.5%
Context window256K
Released
What is Qwen3.5 397B A17B
The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. It delivers state-of-the-art performance comparable to leading-edge models across a wide range of tasks, including language understanding, logical reasoning, code generation, agent-based tasks, image understanding, video understanding, and graphical user interface (GUI) interactions. With its robust code-generation and agent capabilities, the model exhibits strong generalization across diverse agent.
Qwen3.5 397B A17B capabilities
Input: Text、Image、Video Output: Text
Qwen3.5 397B A17B API pricing
Input · per 1M tokens$0.39
Output · per 1M tokens$2.34
USD list-price snapshot, not independently verified; confirm current pricing and terms on the provider site before purchase.
Tools built on this model
通义听悟
Great for meeting-heavy teams, students, journalists, and researchers turning speech into knowledge; not for formal archival use without human proofreading. Verification level not recorded ·