Skip to content

Benchmark

Arena ELO - Russian

People compare model answers and choose which they prefer. Higher scores are better.

Models measured
332
Scores recorded
387
Sources
1
IQ weight
1.42

Score distribution

1,322.5
Mean
1,355.7
Median
134.6
Std dev
1,109.2
10th percentile
1,466.9
90th percentile
936.6
Min
936.61,560.5
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.42 for chat models, measured over 332 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini 4 CarbonHigh· LMArena data1,560.5
2Gemini 4 NeonHigh· LMArena data1,554
3Gemini 4 ArgonHigh· LMArena data1,538.6
4Claude Fable 5.1Max· LMArena data1,518.7
5Gemini 3.8 FlashHigh· LMArena data1,516.3
6Claude Opus 5.5High· LMArena data1,514.4
7Gemini 3.7 FlashHigh· LMArena data1,512.9
8Muse WonderMax· LMArena data1,511.1
9Claude Opus 5Max· LMArena data1,507
10Claude Opus 5High· LMArena data1,506.5
11Claude Fable 5High· LMArena data1,501.4
12Claude Fable 5.2Max· LMArena data1,500.9
13Gemini 3.1 Pro Preview· LMArena data1,497.1
14Muse Spark 1.3Max· LMArena data1,496.4
15Claude Opus 4.6· LMArena data1,495.3
16Claude Opus 4.7High· LMArena data1,495.1
17Claude Opus 4.6High· LMArena data1,494.5
18Gemini 3.5 FlashHigh· LMArena data1,494.3
19Gemini 4 HydrogenHigh· LMArena data1,494
20Gemini 3 Pro Preview· LMArena data1,492.8
21Gemini 3.5 FlashMedium· LMArena data1,492.8
22Gemini 3.6 FlashHigh· LMArena data1,489
23Qwen3.8 Max· LMArena data1,486.7
24Muse Spark 1.2xHigh· LMArena data1,486
25Claude Opus 4.7· LMArena data1,484.9
26Muse Spark 1.1· LMArena data1,483.1
27GPT 5.4High· LMArena data1,480.8
28Kimi K3Max· LMArena data1,480.4
29Gemini 3 Flash· LMArena data1,480.3
30Qwen3.7 Max Preview· LMArena data1,480.2
31MiMo V2.6 Pro· LMArena data1,479.2
32Claude Opus 4.8High· LMArena data1,475.2
33GLM 5.3Max· LMArena data1,474.1
34GPT 5.5· LMArena data1,473.6
35Qwen3.5-Max-Preview· LMArena data1,472.9
36GPT 5.5High· LMArena data1,472.2
37GPT-5.6 SolxHigh· LMArena data1,471.5
38GLM 5.3 Flash· LMArena data1,468.7
39DeepSeek V4.1 Flash 0910Max· LMArena data1,468.5
40Muse Spark· LMArena data1,466.9
41Claude Opus 4.8· LMArena data1,466.3
42GLM 5.2Max· LMArena data1,465.9
43Gemini 2.5 Pro· LMArena data1,460.8
44Gemma 4 31B IT· LMArena data1,459.9
45Claude Sonnet 5.5xHigh· LMArena data1,458.8
46GPT 5.4· LMArena data1,458.8
47ERNIE 5.1· LMArena data1,458.5
48Kimi K3.1 ProMax· LMArena data1,458.5
49Qwen3.7 Plus· LMArena data1,457.8
50Grok 4.20 Multi-Agent0309, beta· LMArena data1,457.5

Showing 1–50 of 387 results.

Data credits: LMArena, CC BY 4.0 (licence)