Skip to content

Benchmark

Arena ELO - Exclude Ties

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.28

Score distribution

1,247.7
Mean
1,316.3
Median
217.2
Std dev
908.5
10th percentile
1,465.2
90th percentile
533.7
Min
533.71,565.6
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.28 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 data1,565.6
2Gemini 4 ArgonHigh· LMArena data1,552.1
3Gemini 4 NeonHigh· LMArena data1,547
4Claude Fable 5.1Max· LMArena data1,532.7
5Claude Opus 5.5High· LMArena data1,532.7
6Claude Opus 5Max· LMArena data1,523.8
7Muse WonderMax· LMArena data1,520
8Claude Opus 5High· LMArena data1,518.2
9Gemini 4 HydrogenHigh· LMArena data1,517.1
10Claude Opus 4.6High· LMArena data1,516.7
11Gemini 3.8 FlashHigh· LMArena data1,515.8
12Claude Opus 4.6· LMArena data1,512
13Muse Spark 1.3Max· LMArena data1,506.4
14Claude Fable 5High· LMArena data1,504.4
15MiMo V2.6 Pro· LMArena data1,504.3
16Claude Opus 4.7High· LMArena data1,501.7
17Gemini 3.7 FlashHigh· LMArena data1,500.8
18Qwen3.8 Max· LMArena data1,496.7
19Muse Spark 1.2xHigh· LMArena data1,493.3
20Claude Opus 4.7· LMArena data1,492.4
21Gemini 3.5 FlashHigh· LMArena data1,491.8
22Gemini 3 Pro Preview· LMArena data1,491
23Gemini 3.1 Pro Preview· LMArena data1,489.6
24Gemini 3.6 FlashHigh· LMArena data1,487.9
25Muse Spark 1.1· LMArena data1,486.8
26Gemini 3.5 FlashMedium· LMArena data1,486
27Claude Fable 5.2Max· LMArena data1,485.9
28GLM 5.3Max· LMArena data1,481.9
29Kimi K3Max· LMArena data1,481.7
30Qwen3.5-Max-Preview· LMArena data1,479.6
31Muse Spark· LMArena data1,479.5
32Qwen3.7 Max Preview· LMArena data1,479.4
33GLM 5.3 Flash· LMArena data1,476.8
34GPT 5.5High· LMArena data1,475.7
35Claude Sonnet 5.5xHigh· LMArena data1,474.7
36GLM 5.2Max· LMArena data1,474.5
37Gemini 3 Flash· LMArena data1,474.1
38GPT 5.4High· LMArena data1,473.5
39ERNIE 5.1· LMArena data1,472
40GPT 5.5· LMArena data1,469.8
41DeepSeek V4.1 Flash 0910Max· LMArena data1,465.8
42MiMo V2.5 Pro· LMArena data1,465.7
43GLM 5.1· LMArena data1,465.2
44Claude Opus 4.8High· LMArena data1,464.1
45Gemini 2.5 Pro· LMArena data1,460.5
46Kimi K3.1 ProMax· LMArena data1,459.2
47GPT-5.6 SolxHigh· LMArena data1,458.2
48Claude Sonnet 4.6· LMArena data1,458.2
49Kimi K2.6· LMArena data1,455.5
50Claude Opus 4.8· LMArena data1,453.9

Showing 1–50 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)