Back to Models

Qwen3 next 80b a3b thinking

Qwen3 Next 80B Thinking is a reasoning-first MoE model (80B total). It specializes in hard multi-step problems and agentic planning.

Thinking Mode
Parameters
80000000000 B
Context
262,144 tokens
Released
Nov 9, 2025

Leaderboards

Performance vs. Industry Average

Intelligence

Qwen3 next 80b a3b thinking is of lower intelligence compared to average (4.1), with an intelligence score of 4.0.

Price

Qwen3 next 80b a3b thinking is cheaper compared to average ($4.91 per 1M Tokens) with a price of $0.75 per 1M Tokens.

Latency

Qwen3 next 80b a3b thinking has a lower average latency compared to average (120.77s), with an average latency of 77.76s.

P99 Latency

Qwen3 next 80b a3b thinking has a lower P99 latency compared to average (354.03s), taking 226.68s to receive the first token at P99 (TTFT).

Context Window

Qwen3 next 80b a3b thinking has a smaller context window than average (347k tokens), with a context window of 262k tokens.

Qwen3 next 80b a3b thinking - AutoBench