Skip to content

Benchmark

Arena ELO - Industry Legal and Government

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

Models measured
338
Scores recorded
394
Sources
1
IQ weight
1.34

Score distribution

1,329.7
Mean
1,368.1
Median
139.4
Std dev
1,104.4
10th percentile
1,475.2
90th percentile
895.4
Min
895.41,543.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.34 for chat models, measured over 338 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 data1,543.1
2Gemini 4 ArgonHigh· LMArena data1,540.7
3Gemini 4 HydrogenHigh· LMArena data1,527.3
4Claude Opus 5Max· LMArena data1,527
5Claude Opus 5.5High· LMArena data1,523
6Claude Fable 5.1Max· LMArena data1,522
7Claude Opus 5High· LMArena data1,521.3
8Gemini 4 NeonHigh· LMArena data1,518.1
9Muse WonderMax· LMArena data1,514.3
10Claude Opus 4.6High· LMArena data1,512.2
11Claude Opus 4.6· LMArena data1,509.5
12Claude Opus 4.7High· LMArena data1,505.3
13Gemini 3.8 FlashHigh· LMArena data1,502.9
14Muse Spark 1.2xHigh· LMArena data1,500.4
15Qwen3.8 Max· LMArena data1,500.4
16Claude Fable 5High· LMArena data1,498.9
17Gemini 3.7 FlashHigh· LMArena data1,497.8
18Kimi K3Max· LMArena data1,496.9
19Muse Spark 1.3Max· LMArena data1,494.7
20GPT 5.5High· LMArena data1,492.4
21Gemini 3 Pro Preview· LMArena data1,491.4
22GLM 5.2Max· LMArena data1,489.9
23Claude Opus 4.7· LMArena data1,488.9
24Gemini 3.1 Pro Preview· LMArena data1,486.7
25Gemini 3.5 FlashHigh· LMArena data1,486.5
26MiMo V2.6 Pro· LMArena data1,486.1
27Gemini 3.5 FlashMedium· LMArena data1,485.6
28Muse Spark· LMArena data1,485.4
29GPT 5.4High· LMArena data1,485.3
30Qwen3.5-Max-Preview· LMArena data1,485
31GLM 5.3Max· LMArena data1,482.2
32GPT 5.5· LMArena data1,482.2
33Muse Spark 1.1· LMArena data1,481.6
34Gemini 2.5 Pro· LMArena data1,480.4
35Gemini 3.6 FlashHigh· LMArena data1,479.3
36GLM 5.3 Flash· LMArena data1,478.9
37Claude Sonnet 5.5xHigh· LMArena data1,477.3
38Claude Opus 4.8· LMArena data1,477
39Claude Opus 4.8High· LMArena data1,475.5
40Qwen3.7 Max Preview· LMArena data1,475.2
41MiMo V2.5 Pro· LMArena data1,474.7
42GPT-5.6 SolxHigh· LMArena data1,474
43GPT-6.1 SolMax· LMArena data1,473.9
44Gemini 3 Flash· LMArena data1,473.3
45Kimi K3.1 ProMax· LMArena data1,472.9
46Claude Fable 5.2Max· LMArena data1,472.3
47Claude Sonnet 4.6· LMArena data1,469.4
48ERNIE 5.1· LMArena data1,468.5
49GLM 5.1· LMArena data1,466.3
50nova-experimental-chat26-02-10· LMArena data1,465.8

Showing 1–50 of 394 results.

Data credits: LMArena, CC BY 4.0 (licence)