Skip to content

Benchmark

Arena ELO - Industry Software and It Services

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.29

Score distribution

1,322.6
Mean
1,374.8
Median
161.3
Std dev
1,081.5
10th percentile
1,486.1
90th percentile
719.3
Min
719.31,552.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.29 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
201Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,383.8
202MiniMax M2.5· LMArena data1,383.3
203Nemotron 3 Nano 30B A3BBF16· LMArena data1,382.3
204Claude Opus 420250514· LMArena data1,381.4
205Grok 3 MiniHigh· LMArena data1,378.8
206DeepSeek V30324· LMArena data1,377.8
207nova-experimental-chat10-09· LMArena data1,376.4
208INTELLECT-3· LMArena data1,375.9
209Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,375.8
210DeepSeek R1· LMArena data1,374.9
211Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,374.8
212Grok 3 Minibeta· LMArena data1,373.2
213MiniMax M2· LMArena data1,372.8
214Qwen2.5-Max· LMArena data1,372.1
215Trinity Large Preview· LMArena data1,371.5
216o4 Mini2025-04-16· LMArena data1,366.7
217o12024-12-17· LMArena data1,366.3
218Ring Flash 2.0· LMArena data1,366
219Qwen3 32B· LMArena data1,365.9
220Granite 4.2 8B· LMArena data1,365.3
221Step3· LMArena data1,365.1
222o3 MiniHigh· LMArena data1,363.7
223Trinity Large Thinking· LMArena data1,362
224O1 Preview· LMArena data1,361.6
225Gemini 2.0 Flash001· LMArena data1,360.9
226Claude Sonnet 420250514· LMArena data1,360.8
227Mistral Small 3.22506· LMArena data1,360.6
228GPT-4.1 mini2025-04-14· LMArena data1,360.1
229Llama 3.3 Nemotron Super 49B v1.5· LMArena data1,359.5
230MiniMax M1· LMArena data1,359.2
231Gemma 3 27B IT· LMArena data1,354.1
232O1 Mini· LMArena data1,353.8
233Step-1o Turbo202506· LMArena data1,353.4
234Hunyuan TurboS20250226· LMArena data1,353
235GLM 4.5V· LMArena data1,352.3
236Olmo 3.1 32B Instruct· LMArena data1,349.3
237Granite 4.2 3B· LMArena data1,349.1
238QwQ 32B· LMArena data1,346.8
239GPT-5 nanoHigh· LMArena data1,345
240o3 Mini· LMArena data1,343.9
241Qwen Plus0125· LMArena data1,342.7
242Qwen3 30B A3B· LMArena data1,339.6
243DeepSeek V3· LMArena data1,336.4
244Command A03-2025· LMArena data1,336.3
245Gemini 2.0 Flash-Lite Preview02-05· LMArena data1,335.1
246Hunyuan Turbo0110· LMArena data1,332
247Olmo 3 32B Think· LMArena data1,331.3
248Llama 3.1 Nemotron Ultra 253B v1· LMArena data1,326.6
249Claude Sonnet 3.7Thinking 32K, 20250219· LMArena data1,323.9
250Granite 4.1 8B· LMArena data1,323

Showing 201–250 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)