Skip to content

Benchmark

Arena ELO - Industry Writing and Literature and Language

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

Models measured
365
Scores recorded
422
Sources
1
IQ weight
1.17

Score distribution

1,297.8
Mean
1,328.7
Median
142.8
Std dev
1,086
10th percentile
1,453.9
90th percentile
809.1
Min
809.11,540.8
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.17 for chat models, measured over 365 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 data1,540.8
2Gemini 4 ArgonHigh· LMArena data1,530
3Claude Opus 5.5High· LMArena data1,528
4Gemini 4 NeonHigh· LMArena data1,517.5
5Claude Fable 5.1Max· LMArena data1,516.1
6Claude Fable 5High· LMArena data1,502.4
7Claude Opus 4.6High· LMArena data1,502
8Gemini 4 HydrogenHigh· LMArena data1,498.9
9Claude Opus 5Max· LMArena data1,498.8
10Claude Fable 5.2Max· LMArena data1,497.6
11Gemini 3.8 FlashHigh· LMArena data1,495.6
12Muse WonderMax· LMArena data1,494.2
13Claude Opus 5High· LMArena data1,493.8
14Claude Opus 4.7High· LMArena data1,493.1
15Claude Opus 4.6· LMArena data1,491.3
16Gemini 3.7 FlashHigh· LMArena data1,489.4
17Gemini 3.1 Pro Preview· LMArena data1,481
18Qwen3.8 Max· LMArena data1,479.9
19Claude Opus 4.7· LMArena data1,479
20Gemini 3.5 FlashHigh· LMArena data1,478.9
21Gemini 3 Pro Preview· LMArena data1,477.7
22Muse Spark 1.3Max· LMArena data1,476.2
23Gemini 3.5 FlashMedium· LMArena data1,473.1
24MiMo V2.6 Pro· LMArena data1,472.1
25Gemini 3.6 FlashHigh· LMArena data1,470.6
26GLM 5.3Max· LMArena data1,468.9
27Claude Sonnet 5.5xHigh· LMArena data1,468.5
28Kimi K3Max· LMArena data1,468.2
29Qwen3.5-Max-Preview· LMArena data1,467.4
30GPT 5.5High· LMArena data1,465
31GPT-5.6 SolxHigh· LMArena data1,464.1
32GPT 5.5· LMArena data1,463.4
33Qwen3.7 Max Preview· LMArena data1,462.7
34Claude Opus 4.8High· LMArena data1,462.2
35GLM 5.2Max· LMArena data1,461.3
36Kimi K3.1 ProMax· LMArena data1,460
37Gemini 3 Flash· LMArena data1,459
38GPT 5.4High· LMArena data1,457
39Muse Spark 1.2xHigh· LMArena data1,456.8
40GLM 5.3 Flash· LMArena data1,456.6
41Muse Spark· LMArena data1,455.1
42GLM 5.1· LMArena data1,454
43Gemini 2.5 Pro· LMArena data1,453.9
44Claude Opus 4.8· LMArena data1,452.7
45MiMo V2.5 Pro· LMArena data1,451.3
46Claude Opus 4.520251101· LMArena data1,451
47ERNIE 5.1· LMArena data1,449.7
48Muse Spark 1.1· LMArena data1,448.7
49GPT 5.4· LMArena data1,448.4
50DeepSeek V4.1 Flash 0910Max· LMArena data1,448.2

Showing 1–50 of 422 results.

Data credits: LMArena, CC BY 4.0 (licence)