Skip to content

Benchmark

Arena ELO Style-Controlled

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.10

Score distribution

1,341.3
Mean
1,363.7
Median
121
Std dev
1,152.6
10th percentile
1,474.5
90th percentile
953.4
Min
953.41,527
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.10 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 data1,527
2Gemini 4 ArgonHigh· LMArena data1,519.6
3Gemini 4 NeonHigh· LMArena data1,512.9
4Muse WonderMax· LMArena data1,511.4
5Claude Opus 5.5High· LMArena data1,506
6Claude Fable 5High· LMArena data1,504.1
7Claude Fable 5.2Max· LMArena data1,503.1
8Claude Opus 4.6High· LMArena data1,502.8
9Claude Fable 5.1Max· LMArena data1,502.8
10Claude Opus 4.7High· LMArena data1,501.2
11Muse Spark 1.3Max· LMArena data1,498.4
12Gemini 3.8 FlashHigh· LMArena data1,497.3
13Claude Opus 4.6· LMArena data1,496.4
14Claude Opus 4.7· LMArena data1,494
15Muse Spark 1.2xHigh· LMArena data1,493.3
16Muse Spark 1.1· LMArena data1,491.1
17Gemini 3.7 FlashHigh· LMArena data1,489.9
18Gemini 4 HydrogenHigh· LMArena data1,489.8
19Claude Opus 5High· LMArena data1,489.7
20Muse Spark· LMArena data1,489.3
21Claude Opus 5Max· LMArena data1,488.8
22Kimi K3Max· LMArena data1,487.8
23Gemini 3.1 Pro Preview· LMArena data1,486.8
24Gemini 3 Pro Preview· LMArena data1,485.6
25GPT-5.6 SolxHigh· LMArena data1,485.1
26Qwen3.8 Max· LMArena data1,484
27Gemini 3.6 FlashHigh· LMArena data1,482.7
28GPT-6.1 SolMax· LMArena data1,482.6
29Kimi K3.1 ProMax· LMArena data1,482.1
30GLM 5.3Max· LMArena data1,481.9
31Claude Opus 4.8High· LMArena data1,481.9
32GPT 5.5High· LMArena data1,481
33MiMo V2.6 Pro· LMArena data1,481
34Gemini 3.5 FlashHigh· LMArena data1,478.5
35GPT 5.5· LMArena data1,476.1
36Gemini 3.5 FlashMedium· LMArena data1,475.9
37GPT-6 AstraMax· LMArena data1,475.3
38GPT-5.2 Chat20260210· LMArena data1,475.3
39GLM 5.3 Flash· LMArena data1,475.3
40GLM 5.2Max· LMArena data1,475.2
41GPT 5.4High· LMArena data1,475.1
42DeepSeek V4.1 Flash 0910Max· LMArena data1,474.6
43Grok 4.20 Beta 1· LMArena data1,474.5
44GPT-6.1 TerraMax· LMArena data1,474.1
45Claude Opus 4.8· LMArena data1,474.1
46Qwen3.7 Max Preview· LMArena data1,473.7
47GPT-5.5 Instant· LMArena data1,473.5
48Claude Opus 4.5High, 32K, 20251101· LMArena data1,473.5
49Claude Sonnet 5.5xHigh· LMArena data1,473.2
50Gemini 3 Flash· LMArena data1,473

Showing 1–50 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)