Qwen

Qwen3.8 2.4T

qwen3.8-2.4t

58Index

Other · restrictedGenerally available984K context2.4T A95BText

Pick this when

Self-hosting Qwen-level quality when you can live with text-only input.

Skip when

You need a permissive license or vision.

Same composite as Max. Different product. Read the license before you resell.

Price

Input / 1M: $1.80
Output / 1M: $5.40
Measured task: $1.09

Downloadable. Text-only. Qwen3.8-Max license restricts some commercial use.

Shape

Context: 984K
Max output: 64K
Released: 2026-08-12
Tools: Yes
Reasoning: Yes

Scores

Intelligence index58

GPQA Diamond89

Terminal-Bench 2.175

Agentic index39

Arena: —
Speed: 55/s
GPQA: 89.1%
Term.: 74.6%

AA Index. The single number most people mean by “how smart.” It is a blend, so a specialist can lose here and still win the job you care about.
GPQA. A clean test of hard reasoning. PhD experts sit around 65%. It says little about writing, tools, or taste.
Term.. Closest public proxy for coding agents that live in a terminal. A high index score with a weak terminal score is a warning.
Arena. Captures taste and usefulness that unit tests miss. Sample size varies by model, so treat gaps under ~100 Elo as noise.

Qwen · Hangzhou. Two flagships: Qwen3.8-Max is a proprietary multimodal API. The 2.4T checkpoint is downloadable, text-only, and commercially restricted.