Qwen3.5-27B: benchmarks, price and rank
Qwen3.5-27B from Alibaba is 82nd of 123 rated AI models on Spezia's leaderboard. It costs $0.26 in, $2.60 out per million tokens.
- 82nd of 123 rated models
- Rating 949
- Rated on 13 benchmarks
- Open weights
- Released Feb 24, 2026
- 262,000 tokens of context
Where it is strongest
Not in the top tenth on any benchmark in use.
Where it falls behind
- 56th of 60 on Vending-Bench 2
- 22nd of 25 on OSWorld-Verified
- 82nd of 98 on LMArena WebDev
Price: $0.26 in, $2.60 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.
- BrowseComp61%
Questions whose answers are hidden deep on the web and take many searches and pages to track down.
37th of 56. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- Humanity's Last Exam48.5%
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.
19th of 74. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- Vending-Bench 2$202
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.
56th of 60. Independent result from Epoch AI.
- OSWorld-Verified56.2%
Ordinary tasks on a real desktop computer using normal apps: a browser, a spreadsheet, an email client, a photo editor, the file manager.
22nd of 25. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- LiveCodeBench v680.7%
Programming contest problems collected after models were trained, from LeetCode, AtCoder and Codeforces.
19th of 43. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- MMMU-Pro75%
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.
33rd of 51. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- CharXiv Reasoning79.5%
Questions about real charts from scientific papers that need reasoning across the chart, not just reading off a value.
28th of 45. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- ScreenSpot-Pro70.3%
Screenshots of professional software at full resolution, with an instruction to find the exact button or field to use.
10th of 18. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- AA-LCR66.1%
Questions that need reasoning across several long documents at once, about 100,000 tokens of text in all.
9th of 19. The maker's own number, from Alibaba Cloud / Qwen Team, via LLM Stats.
- LMArena text1,409
People put a question to two anonymous models and vote for the better answer.
78th of 107. Independent result from LMArena.
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.
Models ranked near it
How the rating works, and every benchmark it uses: the model leaderboard.