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
201GPT OSS 120B· LMArena data1,365.4
202Grok 3 Minibeta· LMArena data1,364.5
203nova-experimental-chat10-09· LMArena data1,364
204Nova 2 Lite· LMArena data1,361.9
205Granite 4.2 30B· LMArena data1,361.1
206Ling Flash 2.0· LMArena data1,361
207MiniMax M2.5· LMArena data1,359.2
208Gemma 3 27B IT· LMArena data1,357.8
209Qwen3 Coder· LMArena data1,356.5
210INTELLECT-3· LMArena data1,356.1
211Mercury 2· LMArena data1,354.8
212Gemini 2.0 Flash001· LMArena data1,354.2
213o4 Mini2025-04-16· LMArena data1,353.1
214O1 Preview· LMArena data1,352.8
215Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,351
216GLM 4.7 Flash· LMArena data1,350.7
217Step3· LMArena data1,350.3
218Nemotron 3 Nano 30B A3BBF16· LMArena data1,347.9
219MiniMax M1· LMArena data1,342.6
220GPT-4.1 mini2025-04-14· LMArena data1,340.5
221MiniMax M2· LMArena data1,340.1
222Qwen3 32B· LMArena data1,340.1
223Trinity Large Thinking· LMArena data1,340
224Claude Sonnet 420250514· LMArena data1,339.2
225Trinity Large Preview· LMArena data1,339
226Mistral Small 3.22506· LMArena data1,338.4
227Llama 3.3 Nemotron Super 49B v1.5· LMArena data1,338.3
228o3 MiniHigh· LMArena data1,336.6
229Step-1o Turbo202506· LMArena data1,335.7
230Gemma 3 12B IT· LMArena data1,334.2
231Ring Flash 2.0· LMArena data1,333.8
232GLM 4.5V· LMArena data1,332.6
233DeepSeek V3· LMArena data1,332.5
234Command A03-2025· LMArena data1,330.9
235GLM 4 Plus0111· LMArena data1,330.9
236Gemini 2.0 Flash-Lite Preview02-05· LMArena data1,329.6
237Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,329.5
238QwQ 32B· LMArena data1,329.1
239Qwen Plus0125· LMArena data1,326.4
240Step-2 16K Exp202412· LMArena data1,320.9
241GPT-5 nanoHigh· LMArena data1,320
242Hunyuan TurboS20250226· LMArena data1,319.8
243o3 Mini· LMArena data1,319.6
244Llama 3.1 Nemotron Ultra 253B v1· LMArena data1,319.5
245Gemini 1.5 Pro002· LMArena data1,319.2
246Granite 4.2 8B· LMArena data1,318.8
247O1 Mini· LMArena data1,317.1
248Qwen3 30B A3B· LMArena data1,316.8
249Claude Sonnet 3.7Thinking 32K, 20250219· LMArena data1,314.4
250Hunyuan Turbo0110· LMArena data1,311.5

Showing 201–250 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)