Skip to content

Benchmark

Arena ELO - Industry Entertainment and Sports and Media

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

Models measured
364
Scores recorded
421
Sources
1
IQ weight
1.17

Score distribution

1,279.8
Mean
1,314.7
Median
149.5
Std dev
1,058.5
10th percentile
1,441.1
90th percentile
784.8
Min
784.81,526.7
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 364 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 data1,431.2
52Claude Sonnet 4.6· LMArena data1,429.9
53Claude Sonnet 4.520250929· LMArena data1,428.5
54Kimi K2.6· LMArena data1,428.2
55DeepSeek-V4-Pro PreviewHigh· LMArena data1,427.3
56Grok 4.20Reasoning, 0309, beta· LMArena data1,426.2
57GLM 5· LMArena data1,425.6
58Claude Opus 4.5High, 32K, 20251101· LMArena data1,425.1
59GPT-6.1 SolMax· LMArena data1,424.1
60DeepSeek V4 ProHigh, 20260813· LMArena data1,423.7
61GPT 5.4· LMArena data1,423.2
62Grok 4.20 Multi-Agent0309, beta· LMArena data1,422.8
63Grok 4.6High· LMArena data1,422.5
64GPT-6 AstraMax· LMArena data1,422.2
65Qwen3.7 Plus· LMArena data1,421.2
66Gemini 3 FlashMinimal· LMArena data1,420.7
67Qwen3.6 Max Preview· LMArena data1,420.5
68Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,419.8
69GPT-5.6 TerraxHigh· LMArena data1,417.4
70Grok 4.20 Beta 1· LMArena data1,416.9
71Kimi K2.5Thinking· LMArena data1,416.7
72Grok 3 Preview02-24· LMArena data1,415.9
73MiMo V2 Pro· LMArena data1,415.6
74Gemini 3.5 Flash Lite· LMArena data1,415
75GPT 5.1High· LMArena data1,414.5
76GPT-6.1 TerraMax· LMArena data1,414.5
77Step 5 Preview· LMArena data1,413.5
78GPT-6.2 AstraMax· LMArena data1,413.4
79MiMo V2.6 Flash· LMArena data1,412.6
80ERNIE 5.00110· LMArena data1,410.7
81Doubao Seed 2.0 Pro· LMArena data1,410.6
82Claude Sonnet 5High· LMArena data1,410.1
83ChatGPT-4o Latest20250326· LMArena data1,409.2
84GLM 4.6· LMArena data1,409
85GPT-6.2 SolMax· LMArena data1,407.8
86Grok 4.1Thinking· LMArena data1,407.4
87DeepSeek V3.2 Exp· LMArena data1,407.4
88GPT-5.5 Instant· LMArena data1,406
89Hy3 (Deep Infra)· LMArena data1,406
90ERNIE-5.0-Preview1203· LMArena data1,405.9
91Grok 4.1· LMArena data1,405.7
92GLM 4.5· LMArena data1,404.6
93Qwen3 Max Preview· LMArena data1,404.5
94Gemma 4 31B IT· LMArena data1,404.2
95Gemma 4 26B A4B· LMArena data1,403.9
96ERNIE-5.0-Preview1022· LMArena data1,403.4
97Qwen3.6 Plus· LMArena data1,401.4
98GPT-5.6 LunaxHigh· LMArena data1,401.2
99DeepSeek R10528· LMArena data1,400.5
100GLM 4.7· LMArena data1,400.1

Showing 51–100 of 421 results.

Data credits: LMArena, CC BY 4.0 (licence)