Skip to content

Benchmark

Arena ELO - Industry Legal and Government

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

Models measured
338
Scores recorded
394
Sources
1
IQ weight
1.34

Score distribution

1,329.7
Mean
1,368.1
Median
139.4
Std dev
1,104.4
10th percentile
1,475.2
90th percentile
895.4
Min
895.41,543.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.34 for chat models, measured over 338 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
201Ling Flash 2.0· LMArena data1,363.4
202nova-experimental-chat10-09· LMArena data1,362.3
203GLM 4 Plus0111· LMArena data1,362.2
204Nova 2 Lite· LMArena data1,358.9
205GLM 4.7 Flash· LMArena data1,358.5
206Qwen3 Coder· LMArena data1,358.2
207Granite 4.2 30B· LMArena data1,355.9
208Qwen 3 235B A22B· LMArena data1,355.7
209Step-1o Turbo202506· LMArena data1,355.2
210GPT OSS 120B· LMArena data1,354.8
211Kimi K2 Preview0711· LMArena data1,353.4
212o4 Mini2025-04-16· LMArena data1,349.9
213Llama 3.3 Nemotron Super 49B v1.5· LMArena data1,348.8
214Qwen Plus0125· LMArena data1,348.7
215MiniMax M2· LMArena data1,348.4
216Command A03-2025· LMArena data1,347.5
217Trinity Large Thinking· LMArena data1,346.5
218O1 Preview· LMArena data1,346.3
219Gemini 2.0 Flash-Lite Preview02-05· LMArena data1,344.5
220GPT-4.1 mini2025-04-14· LMArena data1,343.2
221Nemotron 3 Nano 30B A3BBF16· LMArena data1,341.6
222Claude Sonnet 420250514· LMArena data1,341.5
223DeepSeek V3· LMArena data1,341.4
224Trinity Large Preview· LMArena data1,341.3
225Mistral Small 3.22506· LMArena data1,340.7
226Step3· LMArena data1,339.3
227GLM 4.5V· LMArena data1,337.8
228MiniMax M1· LMArena data1,337.1
229Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,336.7
230Step-2 16K Exp202412· LMArena data1,336.5
231QwQ 32B· LMArena data1,336.4
232Olmo 3.1 32B Instruct· LMArena data1,330.7
233o3 MiniHigh· LMArena data1,328.5
234Gemini 1.5 Pro002· LMArena data1,328.3
235Yi Lightning· LMArena data1,327.7
236Granite 4.2 8B· LMArena data1,327.3
237Gemma 3 4B IT· LMArena data1,327.2
238Granite 4.1 8B· LMArena data1,327.2
239Grok 22024-08-13· LMArena data1,325.7
240Qwen3 32B· LMArena data1,324.6
241INTELLECT-3· LMArena data1,323
242Granite 4.2 3B· LMArena data1,315.4
243O1 Mini· LMArena data1,315.3
244GPT 4o2024-05-13· LMArena data1,313.6
245Qwen3 30B A3B· LMArena data1,313.5
246Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,312.7
247Gemma 3n E4B IT· LMArena data1,309.4
248GLM 4 Plus· LMArena data1,309.2
249Ring Flash 2.0· LMArena data1,307.8
250Qwen2.5-Plus1127· LMArena data1,306.7

Showing 201–250 of 394 results.

Data credits: LMArena, CC BY 4.0 (licence)