MERA - ruHHH: leaderboard
MERA is the Russian open benchmark run by the AI Alliance Russia; this is its ruHHH task: helpful, honest and harmless preference agreement in Russian. Each MERA task keeps its own native metric, so the boards are not on one shared scale.
Metric: Accuracy (%). Source: mera.a-ai.ru. Saturation forecast: Estimated already saturated. 631 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.6 27B | 93.82 | #170 |
| 2 | Kimi K2.5 | 92.13 | #144 |
| 3 | Claude Opus 4.6 | 91.57 | #63 |
| 4 | GLM-5.1 | 91.57 | #97 |
| 5 | Qwen 3.5 27B | 91.01 | #187 |
| 6 | Qwen 3.5 122B A10B | 91.01 | #173 |
| 7 | Gemma 4 31B (IT) | 90.45 | #176 |
| 8 | GPT-5.4 | 89.33 | #79 |
| 9 | Qwen 3 235B A22B 2507 (Thinking) | 89.33 | #260 (Qwen 3 235B A22B 2507) |
| 10 | Qwen 3.6 35B A3B | 89.33 | #226 |
| 11 | Mixtral 8x22B Instruct (v0.1) | 89.33 | #756 |
| 12 | GPT-5.2 | 88.76 | #110 |
| 13 | Qwen 3 30B A3B 2507 (Thinking) | 88.76 | #376 (Qwen 3 30B A3B 2507) |
| 14 | Qwen 3 14B | 88.2 | #537 |
| 15 | gemma-4-E4B-it | 88.2 | #566 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mera-ruhhh · How It Works · Data refreshed daily, snapshot 2026-10-20.