Qwen3 Max Thinking: benchmarks, price and rank
Qwen3 Max Thinking from Alibaba is 91st of 123 rated AI models on Spezia's leaderboard. It costs $1.20 in, $6 out per million tokens.
- 91st of 123 rated models
- Rating 862
- Rated on 11 benchmarks
- Closed
- Released Sep 24, 2025
- 256,000 tokens of context
Where it is strongest
Not in the top tenth on any benchmark in use.
Where it falls behind
- 37th of 37 on Toolathlon
- 74th of 76 on FrontierMath
- 58th of 60 on Vending-Bench 2
Price: $1.20 in, $6 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.
- BrowseComp53.9%
Questions whose answers are hidden deep on the web and take many searches and pages to track down.
41st of 56. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- Vending-Bench 2$72
The model runs a simulated vending machine business for a year on its own: finding suppliers, ordering stock, setting prices and paying the daily fees.
58th of 60. Independent result from Epoch AI.
- Toolathlon18.8%
Long tasks across many real apps and their tools, such as email, calendars, spreadsheets, code repositories and online stores.
37th of 37. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- LiveCodeBench v669%
Programming contest problems collected after models were trained, from LeetCode, AtCoder and Codeforces.
30th of 43. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- SimpleQA Verified48.8%
Short factual questions with one checked answer, asked without web search.
33rd of 78. Independent result from Epoch AI.
- FrontierMath18.9%
Hundreds of new, unpublished math problems written by mathematicians, from hard undergraduate level up to research, with answers a computer can check.
74th of 76. Independent result from Epoch AI.
- AA-LCR68.7%
Questions that need reasoning across several long documents at once, about 100,000 tokens of text in all.
4th of 19. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- LMArena text1,439
People put a question to two anonymous models and vote for the better answer.
52nd of 107. Independent result from LMArena.
- IMO-AnswerBench83.9%
Problems from the International Mathematical Olympiad, the hardest contest for school students, where the model has to reach the right final answer.
9th of 18. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- SWE-bench Verified69.6%
Real bug reports and feature requests from popular open-source Python projects. The model has to change the code so the project's hidden tests pass.
35th of 56. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
4 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.
Models ranked near it
How the rating works, and every benchmark it uses: the model leaderboard.