Qwen3 Max Thinking
Qwen / Qwen
20scores
13benchmarks
$0.78 / $3.9 per 1M tokenscost in/out
Metadata
Qwen Open source
Aliases: qwen-qwen3-max-thinking, qwen-qwen3-max-thinking-20260123, qwen/qwen3-max-thinking, qwen/qwen3-max-thinking-20260123, qwen3-max-thinking, qwen3-max-thinking-20260123
| Benchmark | Category | Rank | Score | Sampled |
|---|---|---|---|---|
| Gert Labs Rankings | Agentic | 33 | 0.47 | 2026-05-11 |
| PinchBench | Agentic | 39 | 0.80 | 2026-05-06 |
| Tau2-Bench Telecom | Agentic | 97 | 83.6% | 2026-05-11 |
| Tau2-Bench Telecom | Agentic | 98 | 83.6% | 2026-05-11 |
| Terminal-Bench Hard | Agentic | 128 | 24.2% | 2026-05-11 |
| Terminal-Bench Hard | Agentic | 165 | 17.4% | 2026-05-11 |
| SciCode | Coding | 57 | 43.1% | 2026-05-11 |
| SciCode | Coding | 128 | 38.7% | 2026-05-11 |
| MageBench Season 1 | Game | 17 | 1594 rating / 12 games | 2026-05-28 |
| Artificial Analysis Intelligence Index | Intelligence | 75 | 39.85 | 2026-05-11 |
| Artificial Analysis Intelligence Index | Intelligence | 127 | 32.48 | 2026-05-11 |
| Humanity's Last Exam | Intelligence | 41 | 26.2% | 2026-05-11 |
| Humanity's Last Exam | Intelligence | 126 | 12% | 2026-05-11 |
| MMLU-Pro | Intelligence | 69 | 82.4% | 2026-05-11 |
| AIME 2025 | Math | 58 | 82.3% | 2026-05-11 |
| Artificial Analysis Openness Index | Openness | 190 | 16.67 | 2026-05-11 |
| GPQA Diamond | Reasoning | 42 | 86.1% | 2026-05-11 |
| GPQA Diamond | Reasoning | 128 | 77.6% | 2026-05-11 |
| CritPt | Science | 63 | 1.7% | 2026-05-11 |
| CritPt | Science | 360 | 0% | 2026-05-11 |
No matching rows.