Skip to content

Benchmark

Arena ELO

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.19

Score distribution

1,312
Mean
1,354.8
Median
146.6
Std dev
1,084.5
10th percentile
1,462.5
90th percentile
833.5
Min
833.51,539.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.19 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
101ChatGPT-4o Latest20250326· LMArena data1,428.5
102DeepSeek R10528· LMArena data1,427.7
103MiMo V2.5· LMArena data1,427.6
104MiniMax M3· LMArena data1,426.5
105LongCat-Flash-Chat Exp2602· LMArena data1,426.2
106GPT-6.1 TerraMax· LMArena data1,426
107Grok 3 Preview02-24· LMArena data1,425.5
108DeepSeek V3.2 ExpThinking· LMArena data1,424.9
109DeepSeek-V4-Flash PreviewHigh· LMArena data1,424.7
110DeepSeek V3.2· LMArena data1,424.6
111Mistral Medium 3.12508· LMArena data1,424.4
112Beam 1.0 501B A23B· LMArena data1,424.3
113MiMo V2 Omni· LMArena data1,422.8
114LongCat Flash Chat· LMArena data1,422.4
115DeepSeek V3.2 Exp· LMArena data1,422.3
116GPT 5.1· LMArena data1,422.2
117Mistral Medium 3.5· LMArena data1,420.7
118Kimi K2.5Instant· LMArena data1,420.4
119DeepSeek V3.2Thinking· LMArena data1,420.1
120Qwen3 VL 235B· LMArena data1,420
121DeepSeek V3.1· LMArena data1,419.8
122GPT-5.5 Instant· LMArena data1,419.4
123DeepSeek V3.1 TerminusThinking· LMArena data1,419.4
124nova-experimental-chat12-10· LMArena data1,419.3
125Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,419.3
126Qwen3 235B A22B Instruct 25072507· LMArena data1,419.2
127Claude Opus 4.120250805· LMArena data1,418.5
128Qwen3.5 122B-A10B· LMArena data1,417.5
129GPT-4.5 Preview2025-02-27· LMArena data1,417.4
130DeepSeek V3.1 Terminus· LMArena data1,417.3
131Qwen3 Next 80B A3B Instruct· LMArena data1,417.2
132Gemini 2.5 Flash· LMArena data1,416.9
133DeepSeek V3.1Thinking· LMArena data1,416.7
134GPT 5.2High· LMArena data1,416.3
135Gemini 3.1 Flash Lite· LMArena data1,416.2
136Qwen 3 235B A22B2507, thinking· LMArena data1,415.4
137nova-experimental-chat11-10· LMArena data1,414.6
138Kimi K2 Thinking· LMArena data1,414.4
139Inkling Small· LMArena data1,414.3
140GPT-6.2 SolMax· LMArena data1,414.3
141GPT-6.2 AstraMax· LMArena data1,414.2
142Mistral Large 4· LMArena data1,413.1
143Qwen3 Max2025-09-23· LMArena data1,412.7
144GPT 5.2· LMArena data1,412
145GPT-5.4 miniHigh· LMArena data1,411.6
146Grok 40709· LMArena data1,411.4
147MiMo V2 FlashNon-Thinking· LMArena data1,411.2
148OpenAI o32025-04-16· LMArena data1,409.7
149Qwen3.5 27B· LMArena data1,408.6
150Grok 4.1 FastReasoning· LMArena data1,408.2

Showing 101–150 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)