Skip to content

Benchmark

Arena ELO - Industry Writing and Literature and Language

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

Models measured
365
Scores recorded
422
Sources
1
IQ weight
1.17

Score distribution

1,297.8
Mean
1,328.7
Median
142.8
Std dev
1,086
10th percentile
1,453.9
90th percentile
809.1
Min
809.11,540.8
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.17 for chat models, measured over 365 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51DeepSeek V4 Pro· LMArena data1,446.7
52Grok 4.5· LMArena data1,445.1
53Claude Opus 4.5High, 32K, 20251101· LMArena data1,444.2
54Claude Sonnet 4.6· LMArena data1,443.5
55GPT-6.1 SolMax· LMArena data1,442.5
56Qwen3.6 Max Preview· LMArena data1,442.4
57DeepSeek V4 ProHigh, 20260813· LMArena data1,442.3
58Claude Sonnet 4.520250929· LMArena data1,441.9
59GPT-6.2 AstraMax· LMArena data1,440.1
60Kimi K2.6· LMArena data1,439.1
61DeepSeek-V4-Pro PreviewHigh· LMArena data1,437.7
62Qwen3.7 Plus· LMArena data1,437.1
63GPT-6 AstraMax· LMArena data1,437
64GLM 5· LMArena data1,436.2
65Gemini 3 FlashMinimal· LMArena data1,436.2
66Claude Sonnet 5High· LMArena data1,434.6
67Grok 4.20Reasoning, 0309, beta· LMArena data1,434.2
68Grok 4.6High· LMArena data1,432.9
69Grok 4.20 Multi-Agent0309, beta· LMArena data1,431.5
70Step 5 Preview· LMArena data1,431.4
71Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,431.2
72GPT-5.6 TerraxHigh· LMArena data1,431.1
73Gemma 4 31B IT· LMArena data1,430.6
74Gemini 3.5 Flash Lite· LMArena data1,427.7
75GPT 5.1High· LMArena data1,427.6
76Grok 4.20 Beta 1· LMArena data1,427.5
77MiMo V2 Pro· LMArena data1,425.2
78Kimi K2.5Thinking· LMArena data1,424.7
79GPT-5.2 Chat20260210· LMArena data1,421.3
80GPT-5.5 Instant· LMArena data1,421.2
81MiMo V2.6 Flash· LMArena data1,421.2
82ERNIE 5.00110· LMArena data1,421.1
83GPT-6.2 SolMax· LMArena data1,420.3
84Hy3 (Deep Infra)· LMArena data1,418.5
85GPT-5.6 LunaxHigh· LMArena data1,418.4
86Qwen3.6 Plus· LMArena data1,418.4
87Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,417.8
88Claude Opus 4.120250805· LMArena data1,417.7
89GPT-6.1 TerraMax· LMArena data1,416.4
90GLM 4.6· LMArena data1,415.7
91Qwen3.5 397B A17B· LMArena data1,415.4
92ChatGPT-4o Latest20250326· LMArena data1,414.7
93ERNIE-5.0-Preview1203· LMArena data1,413.5
94GPT-4.5 Preview2025-02-27· LMArena data1,411.6
95GPT 5.1· LMArena data1,411.3
96Grok 3 Preview02-24· LMArena data1,410.9
97Inkling FP4· LMArena data1,410.8
98Doubao Seed 2.0 Pro· LMArena data1,410.5
99GLM 4.7· LMArena data1,410.5
100DeepSeek V4 Flash· LMArena data1,409.9

Showing 51–100 of 422 results.

Data credits: LMArena, CC BY 4.0 (licence)