MERA - ruTiE: leaderboard
MERA is the Russian open benchmark run by the AI Alliance Russia; this is its ruTiE task: a Russian turn-based interaction evaluation of dialogue consistency. 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. 632 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.6 | 98.09 | #63 |
| 2 | GPT-5.4 | 97.44 | #79 |
| 3 | GLM-5.1 | 96.62 | #97 |
| 4 | Claude Opus 4.5 | 96.38 | #83 |
| 5 | GPT-5.2 | 95.04 | #110 |
| 6 | Qwen 3 235B A22B 2507 (Thinking) | 95.02 | #260 (Qwen 3 235B A22B 2507) |
| 7 | Claude 3.7 Sonnet | 93.27 | #249 |
| 8 | Kimi K2.5 | 92.91 | #144 |
| 9 | Gemini 2.0 Flash | 91.62 | #342 |
| 10 | Qwen 2.5 Max | 91.07 | #416 |
| 11 | Qwen 3 32B | 90.91 | #443 |
| 12 | Qwen 3 30B A3B 2507 (Thinking) | 90.87 | #376 (Qwen 3 30B A3B 2507) |
| 13 | Gemini 1.5 Pro | 90.49 | #445 |
| 14 | GigaChat 2 Max | 90.07 | #536 |
| 15 | DeepSeek V3 | 89.6 | #321 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mera-rutie · How It Works · Data refreshed daily, snapshot 2026-10-20.