Skip to content

Benchmark

Arena ELO - English

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.26

Score distribution

1,327.7
Mean
1,373.1
Median
138.5
Std dev
1,119.5
10th percentile
1,468.7
90th percentile
845.9
Min
845.91,532
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.26 for chat models, measured over 366 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 data1,459.6
52nova-experimental-chat26-02-10· LMArena data1,459.1
53GPT-5.6 SolxHigh· LMArena data1,458.6
54Qwen3.7 Plus· LMArena data1,458.3
55Gemini 2.5 Pro· LMArena data1,458.1
56Grok 4.20Reasoning, 0309, beta· LMArena data1,457.6
57DeepSeek V4 ProHigh, 20260813· LMArena data1,457.5
58GLM 5· LMArena data1,457.4
59Claude Opus 4.5High, 32K, 20251101· LMArena data1,456.4
60Step 5 Preview· LMArena data1,455.7
61GPT-6.1 SolMax· LMArena data1,454.9
62DeepSeek-V4-Pro PreviewHigh· LMArena data1,454.9
63MiMo V2 Pro· LMArena data1,453.8
64Nemotron 3 UltraNVFP4· LMArena data1,453.7
65Kimi K2.5Thinking· LMArena data1,453.4
66Grok 4.5· LMArena data1,453.1
67Grok 4.20 Multi-Agent0309, beta· LMArena data1,452.6
68GLM 4.7· LMArena data1,452.5
69GPT 5.4· LMArena data1,451.6
70Grok 4.20 Beta 1· LMArena data1,451
71GPT 5.1High· LMArena data1,451
72Gemma 4 31B IT· LMArena data1,450.7
73Qwen3.6 Max Preview· LMArena data1,450.4
74LongCat-Flash-Chat Exp2602· LMArena data1,450.4
75Doubao Seed 2.0 Pro· LMArena data1,450.3
76Claude Sonnet 5High· LMArena data1,450.2
77Claude Sonnet 4.520250929· LMArena data1,450.1
78GLM 4.6· LMArena data1,449
79GPT-6 AstraMax· LMArena data1,448.9
80ERNIE 5.00110· LMArena data1,448.4
81Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,448.4
82GPT-5.6 TerraxHigh· LMArena data1,448
83Qwen 3.8 27B· LMArena data1,447.5
84MiMo V2.5· LMArena data1,447
85Hy3 (Deep Infra)· LMArena data1,446.5
86Grok 4.1· LMArena data1,446.4
87GPT-5.2 Chat20260210· LMArena data1,445.9
88Qwen3.6 Plus· LMArena data1,445.1
89Inkling FP4· LMArena data1,444.2
90Gemma 4 26B A4B· LMArena data1,443.8
91ERNIE-5.0-Preview1203· LMArena data1,443.2
92Grok 4.1Thinking· LMArena data1,442.8
93Qwen3.5 397B A17B· LMArena data1,442.4
94Qwen3 Max Preview· LMArena data1,442.4
95DeepSeek V3.2 ExpThinking· LMArena data1,441.4
96GPT-5.6 LunaxHigh· LMArena data1,441.1
97Gemini 3 FlashMinimal· LMArena data1,440.7
98DeepSeek V4 Flash· LMArena data1,440.7
99DeepSeek V3.2 Exp· LMArena data1,440.4
100Mistral Large 3· LMArena data1,439.9

Showing 51–100 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)