Qwen3.6-35B-A3B: benchmarks, price and rank
Qwen3.6-35B-A3B from Alibaba is 101st of 123 rated AI models on Spezia's leaderboard. It costs $0.10 in, $0.95 out per million tokens.
- 101st of 123 rated models
- Rating 772
- Rated on 11 benchmarks
- Open weights
- Released Apr 14, 2026
- 262,000 tokens of context
Where it is strongest
Not in the top tenth on any benchmark in use.
Where it falls behind
- 49th of 50 on SWE-Bench Pro
- 36th of 37 on Toolathlon
- 71st of 76 on FrontierMath
Price: $0.10 in, $0.95 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 Atlas62.8%
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.
24th of 33. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- SWE-Bench Pro49.5%
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.
49th of 50. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- Humanity's Last Exam21.4%
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.
52nd of 74. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- Toolathlon26.9%
Long tasks across many real apps and their tools, such as email, calendars, spreadsheets, code repositories and online stores.
36th of 37. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- LiveCodeBench v680.4%
Programming contest problems collected after models were trained, from LeetCode, AtCoder and Codeforces.
20th of 43. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- FrontierMath20.4%
Hundreds of new, unpublished math problems written by mathematicians, from hard undergraduate level up to research, with answers a computer can check.
71st of 76. Independent result from Epoch AI.
- MMMU-Pro75.3%
College exam questions that need a picture to answer, such as charts, diagrams, maps and chemical structures, with more answer options than the original MMMU and some questions shown only as a screenshot.
31st of 51. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- CharXiv Reasoning78%
Questions about real charts from scientific papers that need reasoning across the chart, not just reading off a value.
31st of 45. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- IMO-AnswerBench78.9%
Problems from the International Mathematical Olympiad, the hardest contest for school students, where the model has to reach the right final answer.
16th of 18. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- SWE-bench Verified73.4%
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.
21st of 56. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
3 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.