Skip to content

Benchmark

Arena ELO - Expert

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

Models measured
318
Scores recorded
371
Sources
1
IQ weight
1.30

Score distribution

1,333.9
Mean
1,371.6
Median
146.5
Std dev
1,113
10th percentile
1,491.8
90th percentile
953.7
Min
953.71,581.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.30 for chat models, measured over 318 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
151Mistral Medium 3.12508· LMArena data1,408.5
152Qwen3.5 35B A3B· LMArena data1,408.4
153GPT-5.5 Instant· LMArena data1,408.1
154Qwen3.5 Flash· LMArena data1,407.6
155nova-experimental-chat12-10· LMArena data1,406
156DeepSeek V3.1· LMArena data1,405.2
157DeepSeek V3.2 Exp· LMArena data1,404.5
158GPT-5 Chat· LMArena data1,404.4
159DeepSeek V3.1Thinking· LMArena data1,404
160OpenAI o32025-04-16· LMArena data1,401.6
161ChatGPT-4o Latest20250326· LMArena data1,400.6
162MiMo V2 FlashThinking· LMArena data1,400.1
163Grok 4.1 FastReasoning· LMArena data1,399.1
164Nemotron 3 Super 120B A12B· LMArena data1,398.8
165Gemini 3.1 Flash Lite· LMArena data1,398.7
166Qwen3 Max2025-09-23· LMArena data1,398.1
167Grok 4 FastNon-Reasoning· LMArena data1,397.1
168Qwen3 30B A3B Instruct2507· LMArena data1,396.6
169GPT-5.3 Chat· LMArena data1,395.7
170Hunyuan T120250711· LMArena data1,394.9
171GPT 5.4 NanoHigh· LMArena data1,394.9
172Grok 3 MiniHigh· LMArena data1,394.6
173GPT-4.5 Preview2025-02-27· LMArena data1,394.2
174DeepSeek R10528· LMArena data1,393.4
175nova-experimental-chat10-20· LMArena data1,386.5
176Muse Glimmer· LMArena data1,386.5
177Claude Opus 4Thinking 16K, 20250514· LMArena data1,385.4
178Grok 4.3· LMArena data1,385.2
179Granite 4.2 8B· LMArena data1,380.4
180GPT-5 miniHigh· LMArena data1,379.1
181MiniMax M2.5· LMArena data1,378.9
182Qwen3 Next 80B A3Bthinking· LMArena data1,375.2
183Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,373.4
184Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,371.7
185GLM 4.6v· LMArena data1,371.6
186GLM 4.5 Air· LMArena data1,369.7
187Claude Opus 420250514· LMArena data1,369.6
188Qwen 3 235B A22BNon-Thinking· LMArena data1,368.7
189Grok 3 Minibeta· LMArena data1,366.5
190Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,366.1
191Kimi K2 Preview0905· LMArena data1,364.9
192GPT 4.12025-04-14· LMArena data1,364.7
193o3 MiniHigh· LMArena data1,364.1
194Qwen3 32B· LMArena data1,362.1
195o12024-12-17· LMArena data1,361.3
196Mercury 2· LMArena data1,361.2
197Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,360.5
198Trinity Large Thinking· LMArena data1,359
199Nova 2 Lite· LMArena data1,358.9
200GLM 4.7 Flash· LMArena data1,358

Showing 151–200 of 371 results.

Data credits: LMArena, CC BY 4.0 (licence)