# Qwen: Qwen3 Max Thinking: AI model fact sheet

- **Provider:** qwen
- **Released:** 2026-02-09
- **Context window:** 262,144 tokens
- **Max output:** 32,768 tokens
- **API pricing:** $1.20 / 1M input, $6.00 / 1M output
- **OpenRouter ID:** qwen/qwen3-max-thinking
- **Capabilities:** conversation, reasoning, code-generation, analysis, tool-use

Qwen3-Max-Thinking is the flagship reasoning model in the Qwen3 series, designed for high-stakes cognitive tasks that require deep, multi-step reasoning. By significantly scaling model capacity and reinforcement learning compute, it delivers major gains in factual accuracy, complex reasoning, instruction following, alignment with human preferences, and agentic behavior. Features Heavy Mode for test-time scaling with iterative refinement, adaptive tool use with integrated search and code interpreter, and hybrid reasoning that toggles between normal and compute-intensive modes mid-conversation.

## Benchmarks

| Benchmark | Score |
| --- | --- |
| GPQA Diamond | 92.8% |
| LiveCodeBench | 91.4% |
| SWE-bench Verified | 75.3% |
| HMMT Feb 25 | 98.0% |
| AIME 2025 | 100% |
| Humanity's Last Exam | 58.3% |
| IMO-AnswerBench | 91.5% |
| Arena-Hard v2 | 90.2% |

Source: real side-by-side outputs, pricing and specs at https://rival.tips/models/qwen3-max-thinking