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
51Gemini 3.5 FlashMedium· LMArena data1,478.4
52GPT-5.6 TerraxHigh· LMArena data1,477.1
53Kimi K2.6· LMArena data1,476.5
54GLM 5.1· LMArena data1,476.4
55Grok 4.6High· LMArena data1,475.8
56Claude Sonnet 5High· LMArena data1,474.8
57DeepSeek V4 Pro· LMArena data1,473.5
58GPT-6 LunaMax· LMArena data1,473.5
59GPT-6.1 TerraMax· LMArena data1,472.6
60Qwen3.6 Max Preview· LMArena data1,471.7
61ERNIE 5.1· LMArena data1,470.8
62GPT 5.1High· LMArena data1,470.6
63Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,470.5
64Step 5 Preview· LMArena data1,470.2
65Muse Spark 1.1· LMArena data1,469.6
66DeepSeek V4 ProHigh, 20260813· LMArena data1,469.5
67GPT-6.2 AstraMax· LMArena data1,469.3
68MiMo V2.6 Flash· LMArena data1,468.6
69GLM 5· LMArena data1,468.5
70MiMo V2 Pro· LMArena data1,467.8
71Claude Opus 4.120250805· LMArena data1,467
72GPT-5.6 LunaxHigh· LMArena data1,466.1
73Qwen3.7 Plus· LMArena data1,466
74Gemini 3 Flash· LMArena data1,465.5
75Gemma 4 31B IT· LMArena data1,464.9
76Gemini 2.5 Pro· LMArena data1,464.4
77GPT-6.2 SolMax· LMArena data1,463.5
78Muse Spark· LMArena data1,462.2
79MiMo V2.5· LMArena data1,461.8
80DeepSeek-V4-Pro PreviewHigh· LMArena data1,461.6
81Qwen3.5 397B A17B· LMArena data1,461.5
82Grok 4.20Reasoning, 0309, beta· LMArena data1,460.8
83Inkling FP4· LMArena data1,460
84Kimi K2.5Thinking· LMArena data1,459.8
85Hy3 (Deep Infra)· LMArena data1,458.7
86GPT 5.1· LMArena data1,457.1
87MiMo V2 Omni· LMArena data1,456.2
88Gemini 3.5 Flash Lite· LMArena data1,456.2
89Qwen 3.8 27B· LMArena data1,455.5
90Grok 4.20 Multi-Agent0309, beta· LMArena data1,455.4
91Qwen3 Max Preview· LMArena data1,455
92DeepSeek V4 Flash· LMArena data1,454.8
93GPT-5.5 Instant· LMArena data1,452.6
94Qwen3.6 Plus· LMArena data1,452.4
95MiniMax M3· LMArena data1,452.3
96Gemini 3 FlashMinimal· LMArena data1,451.8
97GPT-6 SolMax· LMArena data1,451.8
98DeepSeek V3.2· LMArena data1,450.3
99Claude Haiku 4.520251001· LMArena data1,449.6
100Beam 1.0 501B A23B· LMArena data1,449.4

Showing 51–100 of 179 results.

Data credits: LMArena, CC BY 4.0 (licence)