Skip to content

Benchmark

Arena ELO Text Factuality - Longer Query

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

Models measured
160
Scores recorded
179
Sources
1
IQ weight
2.18

Score distribution

1,450.6
Mean
1,455.4
Median
42.5
Std dev
1,387.5
10th percentile
1,494.5
90th percentile
1,307.2
Min
1,307.21,538.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.18 for chat models, measured over 160 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 data1,538.9
2Gemini 4 ArgonHigh· LMArena data1,526.9
3Gemini 4 NeonHigh· LMArena data1,526.8
4Gemini 4 HydrogenHigh· LMArena data1,522.6
5Claude Opus 5.5High· LMArena data1,522.3
6Claude Fable 5.1Max· LMArena data1,519.6
7Claude Opus 4.6High· LMArena data1,513.4
8Claude Opus 4.6· LMArena data1,511.4
9Claude Opus 5Max· LMArena data1,506.4
10Claude Fable 5.2Max· LMArena data1,505.1
11MiMo V2.6 Pro· LMArena data1,505
12Claude Fable 5High· LMArena data1,504.7
13Muse WonderMax· LMArena data1,504.5
14Claude Opus 4.7High· LMArena data1,500.4
15GPT 5.5High· LMArena data1,500.1
16Claude Opus 5High· LMArena data1,498.8
17GPT 5.5· LMArena data1,497.9
18Qwen3.7 Max Preview· LMArena data1,495.6
19Claude Opus 4.7· LMArena data1,494.5
20GPT-5.6 SolxHigh· LMArena data1,494.3
21Qwen3.8 Max· LMArena data1,493.6
22Qwen3.5-Max-Preview· LMArena data1,493.1
23Muse Spark 1.3Max· LMArena data1,493.1
24Claude Opus 4.8High· LMArena data1,492.5
25Gemini 3.8 FlashHigh· LMArena data1,492.1
26Gemini 3.5 FlashHigh· LMArena data1,491.7
27Gemini 3.7 FlashHigh· LMArena data1,491.7
28Claude Opus 4.520251101· LMArena data1,491.5
29GPT 5.4High· LMArena data1,491.3
30Kimi K3Max· LMArena data1,491
31Claude Opus 4.5High, 32K, 20251101· LMArena data1,489.6
32Claude Sonnet 5.5xHigh· LMArena data1,489.5
33Claude Opus 4.8· LMArena data1,487.9
34Kimi K3.1 ProMax· LMArena data1,487.7
35GLM 5.3Max· LMArena data1,487.4
36GPT 5.4· LMArena data1,487.4
37Muse Spark 1.2xHigh· LMArena data1,487.2
38Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,487.1
39Claude Sonnet 4.520250929· LMArena data1,487
40Gemini 3 Pro Preview· LMArena data1,486.1
41MiMo V2.5 Pro· LMArena data1,486.1
42Claude Sonnet 4.6· LMArena data1,485.3
43GLM 5.2Max· LMArena data1,484.6
44DeepSeek V4.1 Flash 0910Max· LMArena data1,484.1
45Gemini 3.1 Pro Preview· LMArena data1,483.4
46GPT-6.1 SolMax· LMArena data1,482.6
47Grok 4.5· LMArena data1,481.1
48Gemini 3.6 FlashHigh· LMArena data1,481
49GLM 5.3 Flash· LMArena data1,480.7
50GPT-6 AstraMax· LMArena data1,480.4

Showing 1–50 of 179 results.

Data credits: LMArena, CC BY 4.0 (licence)