[whatmodel.sh](../../README.md) / Qwen3.7 Max
# Qwen3.7 Max
Alibaba · released 2026-05 · closed weights
**score 72.6** ±0.7 · #54 of 732 · from 16 benchmark results across 6 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Qwen3.7 Max among every model on that task or test, worst to best.
task vs the others rank counts tests
agent tasks ▁▁▃▆█▄▅▅●▄▂▁ #59 of 313 29% 6
coding ▁▁▁▃▅▇█▇▅●▃▁ #52 of 469* 16% 2
math ▁▁▃▅█▆▄█▄●▁▁ #30 of 431 39% 3
web search ▁▁▁▅●█▃▄·▁·▁ #51 of 84* 1.5% 1
agent tasks · #59 of 313 · 29% of its score
Beats [Muse Spark 1.1](muse-spark-1-1.md), [Muse Spark 1.2](muse-spark-1-2.md) here, though they rank higher overall. Trails [Beam 1.0 501B A23B](beam-1-0-501b-a23b.md), [Claude Sonnet 4.6](claude-sonnet-4-6.md), [Kimi K2.6](kimi-k2-6.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Arena Agent (October 2026 aggrega… -0.0659 IPS ▁▁▁▃●▂█▁▂▂▁▁ beats 12/59 avg -0.379 IPS 12% sharp
SWE-bench Verified 77.3% ▁··▁·▁▂▁▄█●▁ beats 24/29 hard · avg 19% 8.0% sharp
Agents' Last Exam overall full-pa… 11.8% ▄···█▄▄●█▄·█ beats 5/10 very hard · avg 1.9% 3.5% typical
MCP-Atlas (public set) 76.4% █··█·██·█●██ beats 5/7 hard · avg 30% 2.5% at its level
Legal Research Bench (agentic, al… 25.2% ▄█··▄▄●██·▄█ beats 5/12 very hard · avg 6.3% 1.5% at its level
EEGAgentBench overall (tool-based… 52.9% ▂·▁▁··▇▇█●·▁ beats 23/25 hard · avg 28% 1.4% blunt · at its level
coding · #52 of 469 · 16% of its score
Beats [GPT-5.4](gpt-5-4.md), [Gemini 3.1 Pro (Preview)](gemini-3-1-pro.md), [DeepSeek V4 Flash Vision (Exp)](deepseek-v4-flash-vision-exp.md) here, though they rank higher overall. Trails [Claude Sonnet 4.6](claude-sonnet-4-6.md), [Qwen3.8 27B](qwen3-8-27b.md), [GLM-5.2](glm-5-2.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Arena WebDev (Code Arena) 1515.1 ▁▁▃▂▅█▃●▅▅▁▂ beats 76/129 avg 981 8.2% sharp
SWE-bench Verified 77.3% ▁··▁·▁▂▁▄█●▁ beats 24/29 hard · avg 19% 8.0% sharp
math · #30 of 431 · 39% of its score
Beats [Gemini 3.8 Flash](gemini-3-8-flash.md), [DeepSeek V4.1 Flash](deepseek-v4-1-flash.md), [Grok 4.6](grok-4-6.md) here, though they rank higher overall.
test result vs the others in its run difficulty counts why
FrontierMath tiers 1–3 v2 64.6% ▂··▁▁▃▄▄●▆▄▃ beats 49/86 very hard · avg 1.2% 22% sharp · at its level
FrontierMath tier 4 v2 34.1% ▁▂▁▂█●▂▃▂▁▁▁ beats 41/67 very hard · avg 0.1% 16% sharp · at its level
OTIS mock AIME 2024–2025 95.6% ▃▃▄▃▃▃▇▇█●▂▅ beats 165/197 very hard · avg 7.2% 0.9% sharp · maxed out · noisy
web search · #51 of 84 · 1.5% of its score
Beats [Gemini 3.1 Pro (Preview)](gemini-3-1-pro.md) here, though they rank higher overall. Trails [GPT-4o mini (2024-07-18)](gpt-4o-mini-2024-07-18.md), [Mistral Small 3.1 24B (2503)](mistral-small-3-1.md), [Llama 4 Scout (109B-A17B)](llama-4-scout.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Legal Research Bench (agentic, al… 25.2% ▄█··▄▄●██·▄█ beats 5/12 very hard · avg 6.3% 1.5% at its level
other tests (6)
test result vs the others in its run difficulty counts why
EBR-bench (Epoch) 9.5% ▄▂·●█▂█▆▄▆·▆ beats 4/25 very hard · avg 0.3% 6.9% sharp
Mystery Game Puzzles (Epoch) 32.0% █··▂▃▇▄●▂▄▂▁ beats 51/77 very hard · avg 9.4% 5.9% sharp · noisy
Chess Puzzles (Epoch) 19.0% █▁▁▁▁▂●▃▃▂▁▁ beats 74/137 very hard · avg 5.8% 3.8% sharp · noisy
GPQA Diamond 90.9% ▁▁▁▁▂▄▆▄▅▇●▅ beats 180/211 hard · avg 39% 3.8% sharp
Legal Research Bench (single-shot… 2.7% ▃●█▃·▅···█▃▃ beats 1/12 very hard · avg 0.8% 1.8% too hard for it
SimpleQA Verified 55.8% ▁▁▃▂▃▃▆▆█●▂▅ beats 61/81 very hard · avg 8.0% 1.8% at its level
source results share of the score
Epoch AI 9 69% ▇▇▇▇▇▇▇▇
LMArena leaderboard dataset 2 20% ▇▇
Agents' Last Exam 1 3% ▇
Legal Research Bench 2 3% ▇
zai-org/GLM-5.2 model card 1 2% ▇
EEGAgentBench: Benchmarking LLM Agents … 1 1% ▇
prior: pull toward the middle 0% ▇
weighted average = 72.6
Licences: Epoch AI ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Agents' Last Exam ([CC BY 4.0](../../licenses/paper-2606.05405v2-CC-BY-4.0.txt)) · Legal Research Bench ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)) · zai-org/GLM-5.2 model card ([mit](../../licenses/zai-org--GLM-5.2.txt)) · EEGAgentBench: Benchmarking LLM Agents on Short- and Long-Horizon EEG Analysis ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/))
## How to run it
### Through an API
api (Deep Infra) · $3.75/M blended · $2.5 in / $7.5 out per 1M tokens · [deepinfra.com](https://deepinfra.com/models)
[price source](https://github.com/anomalyco/models.dev/blob/a668908cfe7073faa0363699486fc41a3ff52412/providers/deepinfra/models/Qwen/Qwen3.7-Max.toml): Standard base token rate from models.dev. Provider-specific tiers, long-context surcharges, caching and promotions may differ. · [context source](https://github.com/anomalyco/models.dev/blob/a668908cfe7073faa0363699486fc41a3ff52412/models/alibaba/qwen3.7-max.toml)
[interactive version](https://whatmodel.sh/model/qwen3-7-max) · [JSON](https://whatmodel.sh/api?format=json&all=1)