Skip to content

Benchmark

Arena ELO - Longer Query

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

Models measured
345
Scores recorded
401
Sources
1
IQ weight
1.34

Score distribution

1,323.7
Mean
1,357.1
Median
137.1
Std dev
1,102.6
10th percentile
1,475.3
90th percentile
951.6
Min
951.61,549
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 345 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 data1,549
2Gemini 4 ArgonHigh· LMArena data1,543.5
3Gemini 4 NeonHigh· LMArena data1,532.8
4Claude Opus 5.5High· LMArena data1,527.6
5Claude Fable 5.1Max· LMArena data1,524.4
6Gemini 4 HydrogenHigh· LMArena data1,520.8
7Claude Opus 4.6High· LMArena data1,517.7
8Claude Opus 5Max· LMArena data1,515.9
9Claude Opus 4.6· LMArena data1,514.9
10Claude Opus 5High· LMArena data1,512
11Claude Fable 5High· LMArena data1,510.2
12Gemini 3.8 FlashHigh· LMArena data1,507.6
13Claude Opus 4.7High· LMArena data1,505.7
14MiMo V2.6 Pro· LMArena data1,504.3
15Claude Opus 4.7· LMArena data1,499.2
16Muse WonderMax· LMArena data1,499
17Claude Fable 5.2Max· LMArena data1,498.9
18Gemini 3.7 FlashHigh· LMArena data1,494.8
19Qwen3.8 Max· LMArena data1,494.5
20Kimi K3Max· LMArena data1,489.8
21Muse Spark 1.3Max· LMArena data1,489
22GLM 5.3Max· LMArena data1,487.3
23Claude Sonnet 5.5xHigh· LMArena data1,484.2
24Claude Opus 4.8High· LMArena data1,484.1
25Gemini 3.5 FlashHigh· LMArena data1,483.7
26GPT 5.5High· LMArena data1,483.3
27GLM 5.3 Flash· LMArena data1,482.3
28GPT 5.5· LMArena data1,482
29Gemini 3.1 Pro Preview· LMArena data1,481.9
30Qwen3.7 Max Preview· LMArena data1,481.6
31MiMo V2.5 Pro· LMArena data1,480.7
32GPT-5.6 SolxHigh· LMArena data1,480.1
33GLM 5.2Max· LMArena data1,479.3
34Claude Opus 4.520251101· LMArena data1,479.2
35Claude Opus 4.8· LMArena data1,478.9
36Claude Sonnet 4.6· LMArena data1,478.5
37Qwen3.5-Max-Preview· LMArena data1,478
38Claude Opus 4.5High, 32K, 20251101· LMArena data1,477.9
39Gemini 3.5 FlashMedium· LMArena data1,477
40Muse Spark 1.2xHigh· LMArena data1,476
41Claude Sonnet 4.520250929· LMArena data1,475.3
42Gemini 3.6 FlashHigh· LMArena data1,474.7
43Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,474.6
44DeepSeek V4.1 Flash 0910Max· LMArena data1,473.7
45GPT 5.4High· LMArena data1,473.1
46Gemini 3 Pro Preview· LMArena data1,471.1
47Kimi K2.6· LMArena data1,468.2
48GLM 5.1· LMArena data1,467.1
49GPT 5.4· LMArena data1,463.8
50Kimi K3.1 ProMax· LMArena data1,463.6

Showing 1–50 of 401 results.

Data credits: LMArena, CC BY 4.0 (licence)