LLM Arena RU: leaderboard
Crowdsourced blind pairwise battles in Russian: Bradley-Terry Elo rating from user votes on llmarena.ru, the Russian-language counterpart of LMArena.
Metric: Arena Elo. Source: llmarena.ru. Saturation forecast: Around March 2027. 112 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 1198 | #66 |
| 2 | Gemini 3.5 Flash | 1191 | #60 |
| 3 | Gemini 2.5 Pro (Preview 03-25) | 1187 | #123 |
| 4 | Gemini 3.1 Pro (Preview) | 1180 | #58 |
| 5 | GPT-5.4 | 1168 | #79 |
| 6 | Qwen 3.6 Max Preview | 1167 | #85 |
| 7 | Claude Opus 4.7 | 1161 | #49 |
| 8 | GPT-5.5 | 1155 | #31 |
| 9 | Gemini 2.5 Flash (Preview 04-17) (Thinking) | 1146 | #237 (Gemini 2.5 Flash (Preview 04-17)) |
| 10 | GPT-5.2 | 1123 | #110 |
| 11 | Claude Sonnet 4.6 | 1120 | #90 |
| 12 | DeepSeek V3.2 | 1120 | #205 |
| 13 | GPT-5 Chat | 1115 | #241 |
| 14 | GPT-5.3 | 1114 | #103 |
| 15 | GPT-5.1 | 1112 | #135 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-arena-ru · How It Works · Data refreshed daily, snapshot 2026-10-20.