Skip to content

Benchmark

Arena ELO - Industry Software and It Services

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.29

Score distribution

1,322.6
Mean
1,374.8
Median
161.3
Std dev
1,081.5
10th percentile
1,486.1
90th percentile
719.3
Min
719.31,552.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.29 for chat models, measured over 366 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 data1,552.3
2Gemini 4 NeonHigh· LMArena data1,541.6
3Gemini 4 ArgonHigh· LMArena data1,541.5
4Claude Opus 5.5High· LMArena data1,533.5
5MiMo V2.6 Pro· LMArena data1,528.7
6Claude Opus 4.6· LMArena data1,526.5
7Claude Opus 5Max· LMArena data1,526.1
8Muse WonderMax· LMArena data1,525.9
9Claude Opus 5High· LMArena data1,525.6
10Claude Opus 4.6High· LMArena data1,525.4
11Claude Fable 5.1Max· LMArena data1,524.7
12Claude Opus 4.7High· LMArena data1,516.1
13Claude Fable 5High· LMArena data1,511.8
14Gemini 4 HydrogenHigh· LMArena data1,511.4
15Muse Spark 1.3Max· LMArena data1,510.4
16Gemini 3.8 FlashHigh· LMArena data1,508.8
17Claude Opus 4.7· LMArena data1,506.7
18Muse Spark 1.2xHigh· LMArena data1,503.9
19Qwen3.8 Max· LMArena data1,502.2
20GLM 5.3 Flash· LMArena data1,500.8
21Gemini 3.7 FlashHigh· LMArena data1,500.2
22Muse Spark 1.1· LMArena data1,498.5
23Kimi K3Max· LMArena data1,497.4
24GLM 5.3Max· LMArena data1,496.1
25DeepSeek V4.1 Flash 0910Max· LMArena data1,495.6
26Claude Sonnet 5.5xHigh· LMArena data1,494.8
27Qwen3.7 Max Preview· LMArena data1,494.5
28MiMo V2.5 Pro· LMArena data1,494.4
29Gemini 3.5 FlashHigh· LMArena data1,494
30Gemini 3.6 FlashHigh· LMArena data1,493.1
31GPT 5.5High· LMArena data1,492.2
32ERNIE 5.1· LMArena data1,492.1
33GPT 5.4High· LMArena data1,491.6
34Claude Sonnet 4.6· LMArena data1,491.1
35MiMo V2.6 Flash· LMArena data1,489.8
36Claude Opus 4.8High· LMArena data1,487.9
37GPT-5.6 SolxHigh· LMArena data1,487.4
38Muse Spark· LMArena data1,487.1
39GLM 5.2Max· LMArena data1,487
40Kimi K3.1 ProMax· LMArena data1,486.6
41Qwen3.5-Max-Preview· LMArena data1,486.6
42GPT 5.5· LMArena data1,486.2
43Gemini 3.5 FlashMedium· LMArena data1,486.1
44Gemini 3.1 Pro Preview· LMArena data1,486
45Gemini 3 Pro Preview· LMArena data1,485.4
46Kimi K2.6· LMArena data1,483.8
47Claude Fable 5.2Max· LMArena data1,483.6
48GLM 5.1· LMArena data1,483.5
49Claude Opus 4.520251101· LMArena data1,483.2
50Claude Opus 4.5High, 32K, 20251101· LMArena data1,482

Showing 1–50 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)