Skip to content

Benchmark

Arena ELO Text Factuality - Chinese

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

Models measured
108
Scores recorded
123
Sources
1
IQ weight
2.06

Score distribution

1,492.7
Mean
1,492
Median
35.3
Std dev
1,451.2
10th percentile
1,534.8
90th percentile
1,389.6
Min
1,389.61,562.4
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.06 for chat models, measured over 108 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 ArgonHigh· LMArena data1,562.4
2Claude Opus 5Max· LMArena data1,555.7
3Claude Opus 5.5High· LMArena data1,554.9
4Claude Fable 5.1Max· LMArena data1,551.3
5Claude Opus 4.6High· LMArena data1,545.4
6Claude Opus 5High· LMArena data1,544
7Claude Opus 4.6· LMArena data1,543.3
8Claude Fable 5High· LMArena data1,540.4
9Claude Sonnet 5.5xHigh· LMArena data1,540
10GPT-5.6 SolxHigh· LMArena data1,538.7
11GPT-6 AstraMax· LMArena data1,537.2
12GPT 5.5· LMArena data1,536
13Kimi K3.1 ProMax· LMArena data1,534.8
14Gemini 3.7 FlashHigh· LMArena data1,534.5
15Gemini 3 Pro Preview· LMArena data1,533.5
16Gemini 3.8 FlashHigh· LMArena data1,532.9
17GPT 5.5High· LMArena data1,532.6
18GPT 5.4High· LMArena data1,530.9
19Muse WonderMax· LMArena data1,530.7
20Qwen3.8 Max· LMArena data1,529.9
21GLM 5.3Max· LMArena data1,528
22Claude Opus 4.7High· LMArena data1,527.9
23Claude Opus 4.7· LMArena data1,527
24GLM 5· LMArena data1,525.5
25GPT 5.4· LMArena data1,524.3
26GPT-6.1 SolMax· LMArena data1,524
27Gemini 3.5 FlashHigh· LMArena data1,524
28GPT-5.6 TerraxHigh· LMArena data1,524
29Gemini 3.6 FlashHigh· LMArena data1,523.6
30Gemini 3.1 Pro Preview· LMArena data1,523.5
31Gemini 3.5 FlashMedium· LMArena data1,522.1
32Muse Spark 1.3Max· LMArena data1,521.4
33Kimi K3Max· LMArena data1,519.7
34Claude Opus 4.8· LMArena data1,518.9
35Gemini 3 Flash· LMArena data1,518
36GLM 5.3 Flash· LMArena data1,517.9
37Kimi K2.6· LMArena data1,517.5
38GLM 4.6· LMArena data1,517.4
39GLM 5.1· LMArena data1,516.2
40GLM 5.2Max· LMArena data1,514.6
41Claude Opus 4.8High· LMArena data1,513.5
42MiMo V2.5 Pro· LMArena data1,512.6
43DeepSeek V4.1 Flash 0910Max· LMArena data1,512.5
44MiMo V2.6 Flash· LMArena data1,511.4
45Muse Spark 1.2xHigh· LMArena data1,509.3
46ERNIE 5.1· LMArena data1,509.1
47GPT 5.1High· LMArena data1,508.9
48Qwen3.7 Plus· LMArena data1,507.4
49Grok 4.5· LMArena data1,505.6
50Muse Spark 1.1· LMArena data1,502.2

Showing 1–50 of 123 results.

Data credits: LMArena, CC BY 4.0 (licence)