Qwen flagship successor to 3.7 Max. AA Intelligence Index 58.1 with a large agentic jump (58.4), at GPQA 92.7.
| Benchmark | Score | Rank |
|---|---|---|
aaAgentic | 58.4% | #3 / 12 |
AA Intelligence Independent composite of agentic work, coding, scientific reasoning, and general capability | 58.1% | #6 / 18 |
LiveCodeBenchvals.ai Contamination-free competitive programming (filtered by cutoff date) | 87.9% | #8 / 59 |
MMMUvals.ai College-level multimodal reasoning across 30+ disciplines | 88% | #10 / 57 |
hleArtificial Analysis | 43% | #11 / 76 |
AA Coding Independent composite of implementation, terminal use, and real-codebase agent performance | 71.8% | #14 / 18 |
GPQAArtificial Analysis PhD-level science questions even experts struggle with | 92.7% | #15 / 88 |
MMLU-Provals.ai Harder 10-option successor to MMLU; more reasoning-focused | 88.6% | #17 / 58 |