Skip to content

Benchmark

Arena ELO - Multi Turn

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

Models measured
364
Scores recorded
421
Sources
1
IQ weight
1.21

Score distribution

1,304.7
Mean
1,350
Median
160.8
Std dev
1,057.9
10th percentile
1,467.2
90th percentile
740.3
Min
740.31,569.6
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.21 for chat models, measured over 364 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
201Qwen 3 235B A22B· LMArena data1,360.5
202Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,359.5
203GLM 4.6v· LMArena data1,358.2
204Mercury 2· LMArena data1,356.1
205Grok 3 MiniHigh· LMArena data1,355.3
206o12024-12-17· LMArena data1,354.7
207nova-experimental-chat10-09· LMArena data1,354.6
208GPT-4.1 mini2025-04-14· LMArena data1,353.8
209Grok 3 Minibeta· LMArena data1,351.1
210INTELLECT-3· LMArena data1,350
211o4 Mini2025-04-16· LMArena data1,349.6
212Gemini 2.0 Flash001· LMArena data1,349.4
213DeepSeek V3· LMArena data1,349
214Qwen3 Next 80B A3Bthinking· LMArena data1,346.3
215Gemma 3 27B IT· LMArena data1,344.9
216Mistral Small 3.22506· LMArena data1,343.5
217GLM 4.7 Flash· LMArena data1,341.6
218Trinity Large Preview· LMArena data1,341.3
219Step-1o Turbo202506· LMArena data1,340.8
220Step3· LMArena data1,340.1
221GPT OSS 120B· LMArena data1,339.3
222Claude Sonnet 3.720250219· LMArena data1,339.1
223Granite 4.2 30B· LMArena data1,339.1
224Command A03-2025· LMArena data1,338.1
225Nova 2 Lite· LMArena data1,338
226Claude Sonnet 3.7Thinking 32K, 20250219· LMArena data1,337.7
227Qwen Plus0125· LMArena data1,336.1
228MiniMax M1· LMArena data1,335.1
229Gemma 3 12B IT· LMArena data1,333.7
230Llama 3.3 Nemotron Super 49B v1.5· LMArena data1,333.4
231GLM 4.5V· LMArena data1,333
232Qwen3 32B· LMArena data1,330.8
233Claude 3.5 Sonnet20241022· LMArena data1,325.8
234Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,324.5
235GLM 4 Plus0111· LMArena data1,323.7
236Ling Flash 2.0· LMArena data1,322.1
237o3 MiniHigh· LMArena data1,320.4
238Trinity Large Thinking· LMArena data1,320.3
239Hunyuan TurboS20250226· LMArena data1,319.5
240Nemotron 3 Nano 30B A3BBF16· LMArena data1,319.2
241Llama 3.1 Nemotron Ultra 253B v1· LMArena data1,317.2
242Llama 3.3 Nemotron Super 49B v1· LMArena data1,315.1
243O1 Mini· LMArena data1,313.9
244QwQ 32B· LMArena data1,313.6
245o3 Mini· LMArena data1,313
246GPT-5 nanoHigh· LMArena data1,311.6
247Yi Lightning· LMArena data1,311.1
248Olmo 3.1 32B Instruct· LMArena data1,310.2
249Gemini 2.0 Flash-Lite Preview02-05· LMArena data1,307.2
250Qwen3 30B A3B· LMArena data1,306.7

Showing 201–250 of 421 results.

Data credits: LMArena, CC BY 4.0 (licence)