Skip to content

Benchmark

Arena ELO Text Factuality - Hard Prompts

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

Models measured
168
Scores recorded
192
Sources
1
IQ weight
2.23

Score distribution

1,459.1
Mean
1,464.1
Median
40.8
Std dev
1,401.2
10th percentile
1,502.4
90th percentile
1,311.6
Min
1,311.61,544.8
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.23 for chat models, measured over 168 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 data1,544.8
2Gemini 4 NeonHigh· LMArena data1,543.1
3Claude Opus 5.5High· LMArena data1,531.5
4Claude Fable 5.1Max· LMArena data1,524.2
5Claude Opus 4.6High· LMArena data1,522.4
6Gemini 4 HydrogenHigh· LMArena data1,520.9
7Claude Opus 4.6· LMArena data1,520.6
8Gemini 4 ArgonHigh· LMArena data1,519.4
9Muse WonderMax· LMArena data1,518.2
10Claude Opus 5Max· LMArena data1,515
11MiMo V2.6 Pro· LMArena data1,513.4
12GPT 5.5High· LMArena data1,512.6
13GPT 5.5· LMArena data1,510.4
14GPT 5.4High· LMArena data1,509.8
15GPT-5.6 SolxHigh· LMArena data1,508.7
16Claude Fable 5High· LMArena data1,508.6
17Claude Opus 5High· LMArena data1,505.4
18Muse Spark 1.3Max· LMArena data1,504.4
19Kimi K3.1 ProMax· LMArena data1,503.9
20Claude Fable 5.2Max· LMArena data1,502.4
21Claude Opus 4.7High· LMArena data1,502.1
22GPT 5.4· LMArena data1,501.5
23Claude Opus 4.7· LMArena data1,501.2
24Qwen3.5-Max-Preview· LMArena data1,500.2
25Muse Spark 1.2xHigh· LMArena data1,499.8
26Claude Opus 4.8High· LMArena data1,499.6
27GPT-6.1 SolMax· LMArena data1,499.6
28Qwen3.7 Max Preview· LMArena data1,498.7
29Claude Sonnet 5.5xHigh· LMArena data1,498.6
30GPT-5.6 TerraxHigh· LMArena data1,498.2
31Kimi K3Max· LMArena data1,498.2
32GPT-6 LunaMax· LMArena data1,497.9
33Gemini 3.5 FlashHigh· LMArena data1,497.6
34Gemini 3 Pro Preview· LMArena data1,497.4
35Qwen3.8 Max· LMArena data1,497.2
36Gemini 3.7 FlashHigh· LMArena data1,497.1
37GPT-6 AstraMax· LMArena data1,496.8
38DeepSeek V4.1 Flash 0910Max· LMArena data1,494.7
39MiMo V2.5 Pro· LMArena data1,493.3
40Claude Sonnet 4.6· LMArena data1,493.1
41GLM 5.3Max· LMArena data1,492.9
42Claude Opus 4.8· LMArena data1,492.4
43Claude Opus 4.5High, 32K, 20251101· LMArena data1,492.2
44Claude Opus 4.520251101· LMArena data1,492.1
45Gemini 3.6 FlashHigh· LMArena data1,492
46Gemini 3.8 FlashHigh· LMArena data1,491.3
47Grok 4.5· LMArena data1,490.6
48Gemini 3.1 Pro Preview· LMArena data1,490.2
49GLM 5.3 Flash· LMArena data1,489
50GPT-5.6 LunaxHigh· LMArena data1,488.5

Showing 1–50 of 192 results.

Data credits: LMArena, CC BY 4.0 (licence)