Skip to content

Benchmark

Arena ELO Text Factuality - Multi Turn

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

Models measured
142
Scores recorded
161
Sources
1
IQ weight
2.11

Score distribution

1,450.4
Mean
1,457.6
Median
36.7
Std dev
1,398.3
10th percentile
1,490.7
90th percentile
1,341.4
Min
1,341.41,524
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.11 for chat models, measured over 142 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 data1,524
2Gemini 4 NeonHigh· LMArena data1,518.6
3Muse WonderMax· LMArena data1,507.9
4Claude Opus 4.6High· LMArena data1,507.3
5Claude Opus 5.5High· LMArena data1,505
6Claude Fable 5High· LMArena data1,503.4
7Claude Opus 4.6· LMArena data1,502.4
8GPT 5.4High· LMArena data1,500
9Claude Opus 4.7High· LMArena data1,497.9
10Claude Opus 5Max· LMArena data1,496.5
11Claude Opus 4.7· LMArena data1,495.7
12GPT 5.5High· LMArena data1,492.9
13Muse Spark 1.2xHigh· LMArena data1,491.6
14GPT 5.5· LMArena data1,491.6
15Gemini 4 HydrogenHigh· LMArena data1,491.6
16Gemini 3 Pro Preview· LMArena data1,491.3
17Kimi K3Max· LMArena data1,490.7
18Claude Opus 4.8High· LMArena data1,490.3
19Muse Spark 1.3Max· LMArena data1,489.9
20Qwen3.8 Max· LMArena data1,489.9
21Gemini 3.5 FlashHigh· LMArena data1,489.6
22Gemini 3.8 FlashHigh· LMArena data1,488.9
23GPT 5.4· LMArena data1,488.8
24Qwen3.5-Max-Preview· LMArena data1,487.2
25Gemini 3.7 FlashHigh· LMArena data1,486.5
26Claude Opus 5High· LMArena data1,486.1
27Claude Fable 5.1Max· LMArena data1,485.9
28GLM 5.3Max· LMArena data1,485
29Gemini 3.6 FlashHigh· LMArena data1,483.8
30Claude Opus 4.8· LMArena data1,483.7
31Gemini 3.1 Pro Preview· LMArena data1,482.4
32GPT-5.6 SolxHigh· LMArena data1,482.3
33GPT-5.6 TerraxHigh· LMArena data1,480.8
34Claude Fable 5.2Max· LMArena data1,480.2
35Muse Spark 1.1· LMArena data1,480
36MiMo V2.5 Pro· LMArena data1,479.8
37Kimi K3.1 ProMax· LMArena data1,479.2
38Claude Opus 4.520251101· LMArena data1,478
39Gemini 3.5 FlashMedium· LMArena data1,478
40Muse Spark· LMArena data1,477.8
41GPT-6.1 SolMax· LMArena data1,477.2
42DeepSeek V4.1 Flash 0910Max· LMArena data1,475.5
43DeepSeek V4 Pro· LMArena data1,475.5
44GLM 5.1· LMArena data1,475.5
45Claude Opus 4.5High, 32K, 20251101· LMArena data1,475.3
46GLM 5.2Max· LMArena data1,475
47Claude Sonnet 5.5xHigh· LMArena data1,475
48Grok 4.5· LMArena data1,474.7
49GPT-6 AstraMax· LMArena data1,473.9
50Claude Sonnet 4.520250929· LMArena data1,471.7

Showing 1–50 of 161 results.

Data credits: LMArena, CC BY 4.0 (licence)