Skip to content

Benchmark

Arena ELO Text Factuality - Industry Writing and Literature and Language

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

Models measured
154
Scores recorded
173
Sources
1
IQ weight
2.19

Score distribution

1,424.1
Mean
1,427.8
Median
44.3
Std dev
1,368.5
10th percentile
1,472.9
90th percentile
1,271.6
Min
1,271.61,520.8
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.19 for chat models, measured over 154 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51GLM 5.3 Flash· LMArena data1,451
52Gemini 2.5 Pro· LMArena data1,449.9
53Claude Sonnet 4.6· LMArena data1,449.5
54ERNIE 5.1· LMArena data1,449.4
55DeepSeek V4 Pro· LMArena data1,448.6
56Muse Spark 1.1· LMArena data1,447.3
57GPT 5.1High· LMArena data1,446.5
58GLM 5· LMArena data1,445.9
59GPT-5.6 TerraxHigh· LMArena data1,445.7
60DeepSeek V4 ProHigh, 20260813· LMArena data1,445.3
61Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,443.7
62Grok 4.20Reasoning, 0309, beta· LMArena data1,443.6
63Kimi K2.6· LMArena data1,443.5
64Grok 4.6High· LMArena data1,443.2
65Gemini 3 FlashMinimal· LMArena data1,442
66Grok 4.20 Multi-Agent0309, beta· LMArena data1,441.9
67DeepSeek-V4-Pro PreviewHigh· LMArena data1,441.8
68GPT-6.2 SolMax· LMArena data1,440.9
69Qwen3.7 Plus· LMArena data1,440.3
70Claude Sonnet 5High· LMArena data1,439.9
71Step 5 Preview· LMArena data1,438.7
72GPT-5.6 LunaxHigh· LMArena data1,438.2
73Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,437.8
74Grok 4.20 Beta 1· LMArena data1,437.4
75GPT-6.1 TerraMax· LMArena data1,436.9
76Gemini 3.5 Flash Lite· LMArena data1,436.8
77GPT-6 LunaMax· LMArena data1,436.6
78Claude Opus 4.120250805· LMArena data1,435
79Gemma 4 31B IT· LMArena data1,434.6
80GPT-5.5 Instant· LMArena data1,433.9
81MiMo V2 Pro· LMArena data1,433.2
82ChatGPT-4o Latest20250326· LMArena data1,432.2
83GPT 5.1· LMArena data1,431.8
84Kimi K2.5Thinking· LMArena data1,430.2
85MiMo V2.6 Flash· LMArena data1,430.1
86GPT-5.2 Chat20260210· LMArena data1,429.6
87ERNIE-5.0-Preview1203· LMArena data1,427.8
88Hy3 (Deep Infra)· LMArena data1,427.2
89Qwen3.5 397B A17B· LMArena data1,426.7
90ERNIE 5.00110· LMArena data1,425.5
91Qwen3.6 Plus· LMArena data1,424.6
92Inkling FP4· LMArena data1,424.1
93DeepSeek V3.2 Exp· LMArena data1,422
94DeepSeek V4 Flash· LMArena data1,422
95Gemini 3.1 Flash Lite· LMArena data1,421.3
96DeepSeek-V4-Flash PreviewHigh· LMArena data1,421.3
97Grok 4.7xHigh· LMArena data1,421.2
98GLM 4.7· LMArena data1,421.1
99GPT 5.2· LMArena data1,421
100MiniMax M3· LMArena data1,420.6

Showing 51–100 of 173 results.

Data credits: LMArena, CC BY 4.0 (licence)