Qwen3.6-Max-Preview is a proprietary frontier model from Alibaba Cloud built on a sparse mixture-of-experts architecture with approximately 1 trillion total parameters. It is optimized for agentic coding, tool use, and...
Tested on 13 benchmarks · BenchGecko score 57.7. Top scores: Chatbot Arena Elo · Coding (1481.5%), Chatbot Arena Elo · Overall (1460.3%), OTIS Mock AIME 2024-2025 (91.1%).
DeepSeek V4.1 Flash scores 57.5 (100% as good) at $0.05/1M input · 95% cheaper
Real-world software engineering tasks from GitHub issues. Models must diagnose bugs and write patches that pass test suites. Human-verified subset of SWE-bench.
Deceptively simple questions that humans find easy but AI models often get wrong. Tests common sense and reasoning gaps.
Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.
Original research-level math problems created by professional mathematicians. Problems are unpublished and cannot be memorized.
Hardest tier of FrontierMath. Problems at the frontier of human mathematical ability, many unsolved by most mathematicians.
- Typetext
- Context262K tokens (~131 books)
- ReleasedApr 2026
- LicenseOpen Source
- Statuspreview
- Cost / Message~$0.008
Frequently Asked Questions
Related Models
Qwen2.5 32B Instruct · AlibabaKimi K2 Thinking · moonshotaiQwen3.8 Max (0902) · Alibaba QwenGrok 4 · xAIDeepSeek V4.1 Flash · DeepSeekKey facts · as of 2026-10-08
- Qwen3.6 Max Preview by Alibaba Qwen. BenchGecko score 57.7, rank 106 of 326 scored models (normalized average of public benchmark scores).
- List price $1.03 input · $6.16 output per 1M tokens (as of 2026-10-08).
How to cite · data as of 2026-10-08
Qwen3.6 Max Preview · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-08. https://benchgecko.ai/model/qwen3-6-max-preview
Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP