Skip to content

Benchmark

Arena ELO Text Factuality - Hard Prompts English

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

Models measured
156
Scores recorded
175
Sources
1
IQ weight
2.08

Score distribution

1,472.5
Mean
1,475.6
Median
36.4
Std dev
1,421.8
10th percentile
1,513.7
90th percentile
1,363.1
Min
1,363.11,545.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.08 for chat models, measured over 156 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 data1,545.9
2Claude Opus 5.5High· LMArena data1,545.3
3Gemini 4 NeonHigh· LMArena data1,536.7
4Claude Opus 4.6High· LMArena data1,533.6
5Gemini 4 ArgonHigh· LMArena data1,532.8
6Muse WonderMax· LMArena data1,532.7
7Claude Fable 5.1Max· LMArena data1,531.9
8Claude Opus 4.6· LMArena data1,531.7
9Claude Opus 5Max· LMArena data1,525.9
10Kimi K3.1 ProMax· LMArena data1,522.7
11Claude Fable 5High· LMArena data1,520.8
12MiMo V2.6 Pro· LMArena data1,520.5
13Claude Opus 5High· LMArena data1,518.1
14Gemini 4 HydrogenHigh· LMArena data1,517.5
15Claude Sonnet 5.5xHigh· LMArena data1,515.2
16GPT 5.5· LMArena data1,514.8
17Qwen3.8 Max· LMArena data1,514.7
18Muse Spark 1.3Max· LMArena data1,513.7
19GPT 5.5High· LMArena data1,513.1
20GPT 5.4High· LMArena data1,512.2
21Claude Opus 4.7High· LMArena data1,511.5
22Kimi K3Max· LMArena data1,510.8
23Claude Fable 5.2Max· LMArena data1,510.5
24Muse Spark 1.2xHigh· LMArena data1,510.2
25Qwen3.5-Max-Preview· LMArena data1,510.2
26Claude Opus 4.7· LMArena data1,510.2
27Claude Opus 4.8High· LMArena data1,510.1
28MiMo V2.5 Pro· LMArena data1,509.6
29GPT-5.6 SolxHigh· LMArena data1,508.7
30GLM 5.3Max· LMArena data1,508.7
31GPT-6.1 SolMax· LMArena data1,508.6
32Claude Sonnet 4.6· LMArena data1,507.4
33DeepSeek V4.1 Flash 0910Max· LMArena data1,504.7
34GPT-6 AstraMax· LMArena data1,503.4
35GPT 5.4· LMArena data1,503.3
36MiMo V2.6 Flash· LMArena data1,502.9
37Claude Opus 4.520251101· LMArena data1,502.9
38Gemini 3.7 FlashHigh· LMArena data1,501.6
39Gemini 3.5 FlashHigh· LMArena data1,501.4
40Gemini 3.8 FlashHigh· LMArena data1,501.2
41GLM 5.3 Flash· LMArena data1,500.9
42Claude Opus 4.8· LMArena data1,500.4
43GPT-5.6 TerraxHigh· LMArena data1,500.3
44Claude Opus 4.5High, 32K, 20251101· LMArena data1,499.5
45Gemini 3.6 FlashHigh· LMArena data1,499.3
46GPT-6.1 TerraMax· LMArena data1,498.2
47Gemini 3 Pro Preview· LMArena data1,498.1
48GLM 5.1· LMArena data1,497.2
49Grok 4.5· LMArena data1,496.8
50Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,496.7

Showing 1–50 of 175 results.

Data credits: LMArena, CC BY 4.0 (licence)