Skip to content

Benchmark

Arena ELO - Chinese

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

Models measured
340
Scores recorded
396
Sources
1
IQ weight
1.23

Score distribution

1,334.6
Mean
1,384.4
Median
172.2
Std dev
1,067.9
10th percentile
1,512.4
90th percentile
719
Min
7191,628.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.23 for chat models, measured over 340 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 HydrogenHigh· LMArena data1,628.9
2Gemini 4 CarbonHigh· LMArena data1,617.3
3Gemini 4 ArgonHigh· LMArena data1,607.7
4Gemini 4 NeonHigh· LMArena data1,598.3
5Claude Fable 5.1Max· LMArena data1,585.9
6Claude Opus 5.5High· LMArena data1,575.3
7Claude Opus 5Max· LMArena data1,575.2
8Claude Opus 5High· LMArena data1,570
9Gemini 3.8 FlashHigh· LMArena data1,561.6
10Gemini 3.7 FlashHigh· LMArena data1,554.4
11Claude Opus 4.6High· LMArena data1,550
12Claude Opus 4.6· LMArena data1,546.3
13Claude Fable 5High· LMArena data1,544.2
14Qwen3.8 Max· LMArena data1,544.1
15Muse WonderMax· LMArena data1,541.8
16GLM 5.3Max· LMArena data1,536.2
17Qwen3.5-Max-Preview· LMArena data1,534.1
18Gemini 3.6 FlashHigh· LMArena data1,533.7
19GPT 5.5· LMArena data1,533.5
20Claude Opus 4.7High· LMArena data1,532.5
21Claude Sonnet 5.5xHigh· LMArena data1,531.9
22MiMo V2.6 Pro· LMArena data1,530.9
23Claude Opus 4.7· LMArena data1,530.1
24Gemini 3.1 Pro Preview· LMArena data1,528.6
25Claude Fable 5.2Max· LMArena data1,528.2
26Kimi K3Max· LMArena data1,528.2
27Muse Spark 1.3Max· LMArena data1,527.7
28Gemini 3.5 FlashMedium· LMArena data1,526.7
29GLM 5.3 Flash· LMArena data1,526.6
30Gemini 3.5 FlashHigh· LMArena data1,525.9
31GPT-5.6 SolxHigh· LMArena data1,525.7
32Qwen3.7 Max Preview· LMArena data1,525.5
33Gemini 3 Pro Preview· LMArena data1,523.7
34GLM 5.2Max· LMArena data1,523.3
35GPT 5.5High· LMArena data1,522.7
36Kimi K2.6· LMArena data1,522
37GPT 5.4High· LMArena data1,517.6
38Muse Spark 1.2xHigh· LMArena data1,517.3
39GLM 5.1· LMArena data1,516.8
40Muse Spark 1.1· LMArena data1,516.7
41GPT-5.6 TerraxHigh· LMArena data1,512.4
42GLM 5· LMArena data1,512
43ERNIE-5.0-Preview1022· LMArena data1,511.9
44MiMo V2.6 Flash· LMArena data1,511.4
45Gemini 3 Flash· LMArena data1,510.3
46GPT 5.4· LMArena data1,509.6
47ERNIE 5.1· LMArena data1,509
48Claude Opus 4.8· LMArena data1,507.9
49Qwen3.7 Plus· LMArena data1,507.3
50Gemini 2.5 Pro· LMArena data1,507.2

Showing 1–50 of 396 results.

Data credits: LMArena, CC BY 4.0 (licence)