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 ·
qwen/qwen3-next-80b-a3b-thinking-2509
Rank #356
qwen
qwen/qwen3-next-80b-a3b-thinking-2509
Usage data date 2026-09-12 · updated daily
Usage and specs
Daily rank#356
Daily tokens153.8M
Daily requests29.8K
Daily change
−2.8%
Released
What is qwen/qwen3-next-80b-a3b-thinking-2509
Qwen3-Next-80B-A3B-Thinking is a reasoning-first chat model in the Qwen3-Next line that outputs structured “thinking” traces by default. It’s designed for hard multi-step problems; math proofs, code synthesis/debugging, logic, and agentic...
qwen/qwen3-next-80b-a3b-thinking-2509 capabilities
Input: Text Output: Text
qwen/qwen3-next-80b-a3b-thinking-2509 API pricing
Input · per 1M tokens$0.15
Output · per 1M tokens$1.20
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 ·