MERA - CheGeKa: leaderboard
MERA is the Russian open benchmark run by the AI Alliance Russia; this is its CheGeKa task: Russian quiz questions from the What? Where? When? tradition, answered as free text. Each MERA task keeps its own native metric, so the boards are not on one shared scale.
Metric: F1 (%). Source: mera.a-ai.ru. Saturation forecast: Around January 2027. 632 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 74.72 | #79 |
| 2 | Claude Opus 4.6 | 74.25 | #63 |
| 3 | Median Human | 71.9 | |
| 4 | Claude Opus 4.5 | 67.89 | #83 |
| 5 | GLM-5.1 | 67.56 | #97 |
| 6 | Claude 3.7 Sonnet | 62.98 | #249 |
| 7 | Gemini 1.5 Pro | 62.98 | #445 |
| 8 | GPT-5.2 | 62.4 | #110 |
| 9 | Kimi K2.5 | 59.11 | #144 |
| 10 | Gemini 2.0 Flash | 58.62 | #342 |
| 11 | DeepSeek V4 Pro | 58.16 | #102 |
| 12 | DeepSeek V4 Flash | 58.16 | #137 |
| 13 | GPT-4o | 55.31 | #343 |
| 14 | DeepSeek V3 (0324) | 53.08 | #340 |
| 15 | DeepSeek V3 | 52.33 | #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-chegeka · How It Works · Data refreshed daily, snapshot 2026-10-20.