PolyMATH
Polymath is a challenging multi-modal mathematical reasoning benchmark designed to evaluate the general cognitive reasoning abilities of Multi-modal Large Language Models (MLLMs). The benchmark comprises 5,000 manually collected high-quality images of cognitive textual and visual challenges across 10 distinct categories, including pattern recognition, spatial reasoning, and relative reasoning.
21rows
scoreprimary metric
2026-05-06sampled
Metadata
Metrics
Score, Normalized Score
| Rank | Subject | Score | Model Match | Provenance | Sampled |
|---|---|---|---|---|---|
| 1 | Qwen3.6 Plus | 0.77 |  Qwen3.6 Plus qwen-qwen3.6-plus | Self-reported | 2026-05-06 |
| 2 | Qwen3.5-397B-A17B | 0.73 | Qwen3.5 397B A17B qwen-qwen3.5-397b-a17b | Self-reported | 2026-05-06 |
| 3 | Qwen3.5-27B | 0.71 | Qwen3.5-27B qwen-qwen3.5-27b | Self-reported | 2026-05-06 |
| 4 | Qwen3.5-122B-A10B | 0.69 | Qwen3.5-122B-A10B qwen-qwen3.5-122b-a10b | Self-reported | 2026-05-06 |
| 5 | Qwen3.5-35B-A3B | 0.64 | Qwen3.5-35B-A3B qwen-qwen3.5-35b-a3b | Self-reported | 2026-05-06 |
| 6 | Qwen3-235B-A22B-Thinking-2507 | 0.60 | Qwen3 235B A22B Thinking 2507 qwen-qwen3-235b-a22b-thinking-2507 | Self-reported | 2026-05-06 |
| 7 | Qwen3.5-9B | 0.57 | Qwen3.5-9B qwen-qwen3.5-9b | Self-reported | 2026-05-06 |
| 8 | Qwen3-Next-80B-A3B-Thinking | 0.56 | Qwen3 Next 80B A3B Thinking qwen-qwen3-next-80b-a3b-thinking | Self-reported | 2026-05-06 |
| 9 | Qwen3 VL 32B Thinking | 0.52 | — | Self-reported | 2026-05-06 |
| 10 | Qwen3 VL 30B A3B Thinking | 0.52 | Qwen3 VL 30B A3B Thinking qwen-qwen3-vl-30b-a3b-thinking | Self-reported | 2026-05-06 |
| 11 | Qwen3.5-4B | 0.51 | — | Self-reported | 2026-05-06 |
| 12 | Qwen3-235B-A22B-Instruct-2507 | 0.50 | Qwen3 235B A22B Instruct 2507 qwen-qwen3-235b-a22b-2507 | Self-reported | 2026-05-06 |
| 13 | Qwen3 VL 8B Thinking | 0.47 | Qwen3 VL 8B Thinking qwen-qwen3-vl-8b-thinking | Self-reported | 2026-05-06 |
| 14 | Qwen3-Next-80B-A3B-Instruct | 0.46 | Qwen3 Next 80B A3B Instruct qwen-qwen3-next-80b-a3b-instruct | Self-reported | 2026-05-06 |
| 15 | Qwen3 VL 4B Thinking | 0.45 | — | Self-reported | 2026-05-06 |
| 16 | Qwen3 VL 30B A3B Instruct | 0.44 | Qwen3 VL 30B A3B Instruct qwen-qwen3-vl-30b-a3b-instruct | Self-reported | 2026-05-06 |
| 17 | Qwen3 VL 32B Instruct | 0.41 | Qwen3 VL 32B Instruct qwen-qwen3-vl-32b-instruct | Self-reported | 2026-05-06 |
| 18 | Qwen3 VL 8B Instruct | 0.30 | Qwen3 VL 8B Instruct qwen-qwen3-vl-8b-instruct | Self-reported | 2026-05-06 |
| 19 | Qwen3 VL 4B Instruct | 0.29 | — | Self-reported | 2026-05-06 |
| 20 | Qwen3.5-2B | 0.26 | — | Self-reported | 2026-05-06 |
| 21 | Qwen3.5-0.8B | 0.08 | — | Self-reported | 2026-05-06 |
No matching rows.