MERA v2: leaderboard
The 2.0 text suite that replaced MERA v1 as the default board at mera.a-ai.ru: twelve newly written Russian tasks (SAGE, Humor, LIMUR, RUBIN, Riddles, SOB-Hard, Characters, Enantiosemy and four more) scored 0-1 as a total over the set. Answers stay private and submissions are run by the organisers, so the task mix differs from v1 and the two scores are not comparable.
Metric: Total score. Source: mera.a-ai.ru. Saturation forecast: Around October 2027. 40 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.7 Flash | 0.68 | #23 |
| 2 | GPT-5.6 Sol | 0.64 | #19 |
| 3 | Claude Opus 5 | 0.63 | #12 |
| 4 | Grok 4.6 | 0.62 | #26 |
| 5 | GPT-5.6 Terra | 0.57 | #41 |
| 6 | GPT-5.6 Luna | 0.52 | #68 |
| 7 | Qwen 3.5 397B A17B | 0.48 | #146 |
| 8 | Grok 4.3 | 0.47 | #188 |
| 9 | MiMo-V2.5-Pro | 0.44 | #158 |
| 10 | Qwen 3.5 122B A10B | 0.43 | #173 |
| 11 | Qwen 3.6 27B | 0.43 | #170 |
| 12 | Qwen 3.5 27B | 0.42 | #187 |
| 13 | GigaChat 3.5 Reasoning | 0.42 | #167 |
| 14 | Qwen 3.7 Flash | 0.41 | #171 |
| 15 | MiniMax-M3 | 0.4 | #136 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mera-v2 · How It Works · Data refreshed daily, snapshot 2026-10-20.