[whatmodel.sh](../../README.md) / Qwen3-Max (2025-09-23) # Qwen3-Max (2025-09-23) Alibaba · released 2025-09 · proprietary **score 65.1** ±1.0 · #141 of 732 · from 12 benchmark results across 4 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at _{● marks Qwen3-Max (2025-09-23) among every model on that task or test, worst to best.}

task            vs the others         rank  counts  tests
agent tasks     ▁▁▃▆█▄●▅▄▄▂▁   #124 of 313     11%      3
chat & writing  ▁▁▁▃▆▄▆█▆●▇▃   #89 of 402*     10%      1
math            ▁▁▃▅█▆▄●▄▂▁▁   #118 of 431     47%      3

agent tasks · #124 of 313 · 11% of its score

Beats [o3-pro](o3-pro.md), [o3](o3.md), [GPT-5 mini](gpt-5-mini.md) here, though they rank higher overall. Trails [SmolLM3 3B](smollm3-3b.md), [Jamba 1.5 Large](jamba-1-5-large.md), [Devstral 2 (2512)](devstral-2.md) here, though they rank lower overall.

test                      result  vs the others in its run  difficulty        counts  why
τ²-bench Telecom (maker)   84.2%  ▁·▅█▆▅▂●▃▂▃▁ beats 62/89  medium · avg 37%    5.4%  sharp · small run
τ²-bench Retail (maker)    72.2%  ▁····▂▁▃▆●▃▁ beats 33/44  medium · avg 39%    3.3%  at its level · small run
τ²-bench Airline (maker)   59.5%  ▁▁·▂▁▃▂▁█●▂▃ beats 30/42  hard · avg 30%      2.7%  at its level · small run

chat & writing · #89 of 402 · 10% of its score

Beats [LongCat 2.0 (1.6T-A48B)](longcat-2-0.md), [Inkling Small (266B-A12B)](inkling-small.md), [Qwen3.5 122B-A10B](qwen3-5-122b-a10b.md) here, though they rank higher overall. Trails [Gemma 4 12B](gemma-4-12b.md), [Gemma 4 26B-A4B](gemma-4-26b-a4b.md), [Mistral Medium 3.5 (128B)](mistral-medium-3-5.md) here, though they rank lower overall.

test            result  vs the others in its run    difficulty  counts  why
LMArena (text)  1423.5  ▁▁▁▂▃▃▃▆▆●█▃ beats 276/385  avg 1260       10%  blunt

math · #118 of 431 · 47% of its score

Beats [Claude Sonnet 4.6](claude-sonnet-4-6.md), [MiniMax-M3](minimax-m3.md), [MiMo-V2.5 (310B-A15B)](mimo-v2-5.md) here, though they rank higher overall. Trails [Olmo 3.1 32B Think](olmo-3-1-32b-think.md), [gpt-oss-120b](gpt-oss-120b.md), [Qwen3-VL 32B Thinking](qwen3-vl-32b-thinking.md) here, though they rank lower overall.

test                       result  vs the others in its run    difficulty            counts  why
FrontierMath tiers 1–3 v2   18.9%  ▂··▁▁●▄▄█▆▄▃ beats 12/86    very hard · avg 1.2%     33%  sharp
MATH level 5                97.1%  ▂▂▅▆▄█▆▃▅▂▃● beats 87/92    hard · avg 28%          7.5%  sharp
OTIS mock AIME 2024–2025    73.3%  ▃▃▄▃▃▃▇●█▃▂▅ beats 107/197  very hard · avg 7.2%    6.4%  sharp · at its level · noisy

other tests (5)

test                                result  vs the others in its run    difficulty            counts  why
GPQA Diamond                         72.6%  ▁▁▁▁▂▄▆▄●▇█▅ beats 110/211  hard · avg 39%           15%  sharp · at its level
KINA (Knowledge Index of Noah's A…   35.9%  ▁▁▂▁▃▄█▆●▅▂▃ beats 29/39    very hard · avg 21%     6.8%  at its level
Chess Puzzles (Epoch)                 4.0%  ●▁▁▁▁▂▃▃▃▂▁▁ beats 1/137    very hard · avg 5.8%    3.5%  sharp · noisy
Mystery Game Puzzles (Epoch)          5.0%  ●··▂▃▇▄█▂▄▂▁ beats 1/77     very hard · avg 9.4%    3.5%  sharp · too hard for it · noisy
SimpleQA Verified                    48.7%  ▁▁▃▂▃▃▆▆●▄▂▅ beats 52/81    very hard · avg 8.0%    3.1%  at its level

## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its *counts* column above; here it is added up by who ran the tests. [Method](../method.md)

source                                    results  share of the score
Epoch AI                                        7   72% ▇▇▇▇▇▇▇▇▇
τ²-bench leaderboard: Qwen (submitted /…        3   11% ▇
LMArena leaderboard dataset                     1   10% ▇
Knowledge Index of Noah's Ark                   1    7% ▇
prior: pull toward the middle                        0% ▇
weighted average                                   = 65.1

_{Licences: Epoch AI ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · τ²-bench leaderboard: Qwen ([MIT](https://opensource.org/license/mit)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Knowledge Index of Noah's Ark ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/))} ## How to run it ### Through an API api (Alibaba Cloud Model Studio) · $2.40/M blended · $1.2 in / $6 out per 1M tokens · [alibabacloud.com](https://www.alibabacloud.com/help/en/model-studio/text-generation-model) _{[price source](https://www.alibabacloud.com/help/en/model-studio/model-pricing): USD per 1M tokens, Singapore (International), dated snapshot qwen3-max-2025-09-23 (non-thinking mode only), 0\

[interactive version](https://whatmodel.sh/model/qwen3-max-2025-09-23) · [JSON](https://whatmodel.sh/api?format=json&all=1)}