Skip to content
← Every AI model, ranked
Alibaba

Qwen3.7 Max: benchmarks, price and rank

Qwen3.7 Max from Alibaba is 40th of 123 rated AI models on Spezia's leaderboard, strongest on LiveCodeBench v6 and IMO-AnswerBench. It costs $1.25 in, $3.75 out per million tokens.

  • 40th of 123 rated models
  • Rating 1,418
  • First on 1 of 12 benchmarks
  • Closed
  • Released May 19, 2026
  • 1 million tokens of context
Compare Qwen3.7 Max with any model Free account. No card.

Where it is strongest

  • 3rd of 43 on LiveCodeBench v6
  • 3rd of 18 on IMO-AnswerBench
  • 1st of 56 on MMLU-Pro

Where it falls behind

  • 39th of 76 on FrontierMath

Price: $1.25 in, $3.75 out per million tokens.

Its headline results

The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.

  • MCP Atlas76.4%

    Multi-step tasks where the model has to pick and combine many real tools, connected through MCP, the standard way agents plug into apps and data.

    10th of 33. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.

  • SWE-Bench Pro60.6%

    Long changes to real codebases in several languages, often across many files, including code from private and copyleft projects that models are unlikely to have seen.

    19th of 50. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.

  • Humanity's Last Exam41.4%

    About 2,500 questions written by experts in more than a hundred fields to be as hard as they could make them, many needing a diagram or image.

    27th of 74. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.

  • LiveCodeBench v691.6%

    Programming contest problems collected after models were trained, from LeetCode, AtCoder and Codeforces.

    3rd of 43. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.

  • SimpleBench70.4%

    Multiple-choice puzzles about everyday situations, space, time and people, written to trip up models that pattern-match instead of think. Ordinary adults do well on them.

    24th of 68. Independent result from Epoch AI.

  • SimpleQA Verified55.8%

    Short factual questions with one checked answer, asked without web search.

    25th of 78. Independent result from Epoch AI.

  • FrontierMath64.6%

    Hundreds of new, unpublished math problems written by mathematicians, from hard undergraduate level up to research, with answers a computer can check.

    39th of 76. Independent result from Epoch AI.

  • LMArena text1,475

    People put a question to two anonymous models and vote for the better answer.

    21st of 107. Independent result from LMArena.

  • LMArena WebDev1,515

    People ask two anonymous models to build the same web page or app and vote for the better one.

    49th of 98. Independent result from LMArena.

  • IMO-AnswerBench90%

    Problems from the International Mathematical Olympiad, the hardest contest for school students, where the model has to reach the right final answer.

    3rd of 18. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.

7 more results in its full profile.

Every result with its source, what it costs per task, what jumped with its release, side by side with any other model, and answers to your questions about when to use it.

See the full profile

Models ranked near it

How the rating works, and every benchmark it uses: the model leaderboard.