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
101MiMo V2 Omni· LMArena data1,439.4
102DeepSeek V3.2Thinking· LMArena data1,439.4
103MiniMax M3· LMArena data1,439.3
104LongCat Flash Chat· LMArena data1,438.9
105nova-experimental-chat12-10· LMArena data1,438.1
106Mistral Medium 3.12508· LMArena data1,438
107Grok 3 Preview02-24· LMArena data1,437.7
108Gemini 3.5 Flash Lite· LMArena data1,437.7
109DeepSeek R10528· LMArena data1,437.2
110GLM 4.5· LMArena data1,437
111DeepSeek V3.2· LMArena data1,436.9
112Beam 1.0 501B A23B· LMArena data1,435.5
113DeepSeek-V4-Flash PreviewHigh· LMArena data1,435.1
114ChatGPT-4o Latest20250326· LMArena data1,434.5
115Kimi K2.5Instant· LMArena data1,433.1
116Grok 4.6High· LMArena data1,433.1
117Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,432.8
118Mistral Medium 3.5· LMArena data1,432.6
119DeepSeek V3.1Thinking· LMArena data1,432.2
120GPT-6.1 TerraMax· LMArena data1,431.3
121NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,430.9
122DeepSeek V3.1· LMArena data1,430.1
123Qwen 3 235B A22B2507, thinking· LMArena data1,430
124MiMo V2 FlashNon-Thinking· LMArena data1,429.9
125Qwen3 VL 235B· LMArena data1,429.1
126Qwen3.5 122B-A10B· LMArena data1,429
127DeepSeek V3.1 TerminusThinking· LMArena data1,428.7
128ERNIE-5.0-Preview1022· LMArena data1,428.5
129Hunyuan Vision 1.5 Thinking· LMArena data1,426.8
130Claude Opus 4.120250805· LMArena data1,426.5
131Kimi K2 Thinking· LMArena data1,426.4
132GPT 5.1· LMArena data1,425.5
133MiniMax M2.7· LMArena data1,425.3
134Qwen3.5 27B· LMArena data1,424.7
135Qwen3 235B A22B Instruct 25072507· LMArena data1,424
136Mistral Large 4· LMArena data1,422.7
137GPT 5.2High· LMArena data1,422.3
138Qwen3 Next 80B A3B Instruct· LMArena data1,421.7
139nova-experimental-chat11-10· LMArena data1,420.8
140Inkling Small· LMArena data1,420.6
141Grok 4.1 FastReasoning· LMArena data1,420.2
142GPT-4.5 Preview2025-02-27· LMArena data1,419.7
143Qwen3 VL 235Bthinking· LMArena data1,419.6
144GPT-6.2 SolMax· LMArena data1,419.4
145DeepSeek V3.1 Terminus· LMArena data1,418.5
146Qwen3 Max2025-09-23· LMArena data1,418.2
147Grok 40709· LMArena data1,418.1
148Inkling 2 Small· LMArena data1,417.8
149Gemini 2.5 Flash· LMArena data1,417.7
150GPT-5.5 Instant· LMArena data1,417.6

Showing 101–150 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)