MERA Reason - ruAIME: leaderboard
MERA Reason is the Russian reasoning suite MERA launched on 2026-07-24, run as its own Hugging Face Space; this board is a Russian AIME-style competition-mathematics set.
Metric: Score. Source: huggingface.co. Saturation forecast: Around April 2027. 24 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.8 | 98.48 | #37 |
| 2 | Gemma 4 31B (IT) | 95.03 | #176 |
| 3 | Qwen 3.6 27B | 91.71 | #170 |
| 4 | GLM-5 | 90.06 | #141 |
| 5 | Qwen 3.5 35B A3B | 88.81 | #256 |
| 6 | DeepSeek V4 Pro | 88.67 | #102 |
| 7 | Qwen 3.6 35B A3B | 88.54 | #226 |
| 8 | Qwen 3 30B A3B 2507 (Thinking) | 87.85 | #376 (Qwen 3 30B A3B 2507) |
| 9 | Qwen 3.5 27B | 87.71 | #187 |
| 10 | DeepSeek V4 Flash | 86.88 | #137 |
| 11 | Qwen 3.5 122B A10B | 86.88 | #173 |
| 12 | Qwen 3 Next 80B A3B (Thinking) | 84.53 | #320 (Qwen 3 Next 80B A3B) |
| 13 | Qwen 3.5 9B | 78.73 | #379 |
| 14 | Qwen 3 4B 2507 (Thinking) | 77.76 | #538 (Qwen 3 4B 2507) |
| 15 | Gemma 4 26B A4B (IT) | 77.62 | #297 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mera-reason-ruaime · How It Works · Data refreshed daily, snapshot 2026-10-20.