Skip to content

Benchmark

Arena ELO Text Factuality - Russian

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

Models measured
130
Scores recorded
146
Sources
1
IQ weight
2.22

Score distribution

1,450.9
Mean
1,456.3
Median
39.2
Std dev
1,407.8
10th percentile
1,497.3
90th percentile
1,300.5
Min
1,300.51,534.6
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.22 for chat models, measured over 130 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 NeonHigh· LMArena data1,534.6
2Claude Fable 5.2Max· LMArena data1,520.8
3Gemini 4 ArgonHigh· LMArena data1,519.5
4Claude Opus 5.5High· LMArena data1,518.1
5Muse WonderMax· LMArena data1,516.8
6Claude Fable 5.1Max· LMArena data1,513.5
7Claude Fable 5High· LMArena data1,506.6
8Gemini 3.7 FlashHigh· LMArena data1,506.4
9Claude Opus 5Max· LMArena data1,505.1
10Claude Opus 5High· LMArena data1,500.5
11Gemini 3.8 FlashHigh· LMArena data1,499.9
12Claude Opus 4.6· LMArena data1,498.9
13Gemini 3.5 FlashHigh· LMArena data1,498.8
14GPT 5.4High· LMArena data1,498.5
15Gemini 3.1 Pro Preview· LMArena data1,497.3
16Muse Spark 1.3Max· LMArena data1,496
17Gemini 4 HydrogenHigh· LMArena data1,495.4
18Claude Opus 4.6High· LMArena data1,495.2
19Claude Opus 4.7High· LMArena data1,494.8
20GPT 5.5· LMArena data1,492.4
21Gemini 3 Pro Preview· LMArena data1,491.8
22Qwen3.5-Max-Preview· LMArena data1,490.5
23GPT-5.6 SolxHigh· LMArena data1,489.2
24Claude Opus 4.7· LMArena data1,489.2
25Gemini 3.6 FlashHigh· LMArena data1,489.2
26Qwen3.8 Max· LMArena data1,488.7
27Gemini 3.5 FlashMedium· LMArena data1,488.6
28Kimi K3.1 ProMax· LMArena data1,488.3
29GPT 5.5High· LMArena data1,486.5
30Claude Opus 4.8High· LMArena data1,485.1
31GPT-6.1 SolMax· LMArena data1,485
32MiMo V2.6 Pro· LMArena data1,484.8
33Kimi K3Max· LMArena data1,483.9
34Muse Spark 1.2xHigh· LMArena data1,483.3
35GLM 5.3Max· LMArena data1,482
36Gemini 3 Flash· LMArena data1,480.5
37GPT-6 AstraMax· LMArena data1,480
38Muse Spark 1.1· LMArena data1,479.5
39GPT-5.6 TerraxHigh· LMArena data1,479.4
40Grok 4.5· LMArena data1,477.2
41DeepSeek V4.1 Flash 0910Max· LMArena data1,476.9
42GPT 5.4· LMArena data1,476.3
43Claude Opus 4.8· LMArena data1,474
44GLM 5.3 Flash· LMArena data1,471.4
45Claude Sonnet 5.5xHigh· LMArena data1,471.2
46GLM 5.2Max· LMArena data1,470.7
47Qwen3.7 Plus· LMArena data1,468.4
48Grok 4.20Reasoning, 0309, beta· LMArena data1,468.4
49Gemini 2.5 Pro· LMArena data1,468.4
50Gemini 3 FlashMinimal· LMArena data1,466.2

Showing 1–50 of 146 results.

Data credits: LMArena, CC BY 4.0 (licence)