MATH-MC Level 4: leaderboard
MATH-MC Level 4 evaluates multiple-choice mathematical reasoning on advanced MATH problems.
Metric: Accuracy (%). Source: github.com. Saturation forecast: Estimated already saturated. 69 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 99.58 | #58 |
| 2 | Kimi K2.5 | 99.58 | #144 |
| 3 | Gemma 4 31B (IT) | 99.58 | #176 |
| 4 | Qwen 3.5 122B A10B | 99.58 | #173 |
| 5 | MiMo-V2-Flash | 99.58 | #372 |
| 6 | GLM-5.1 FP8 | 99.58 | #109 |
| 7 | GPT-5.5 | 99.5 | #31 |
| 8 | Claude Opus 4.6 | 99.5 | #63 |
| 9 | NVIDIA-Nemotron-3-Super-120B-A12B-FP8 | 99.5 | #223 |
| 10 | Qwen 3.5 35B A3B | 99.42 | #256 |
| 11 | Qwen 3 Next 80B A3B (Thinking) | 99.42 | #320 (Qwen 3 Next 80B A3B) |
| 12 | Step 3.5 Flash | 99.42 | #270 |
| 13 | DeepSeek V3.2 Speciale | 99.42 | #128 |
| 14 | GLM-5 FP8 | 99.42 | #143 |
| 15 | Qwen 3.5 397B A17B | 99.33 | #146 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=math-mc-level-4 · How It Works · Data refreshed daily, snapshot 2026-10-20.