Skip to content

Benchmark

Arena ELO - Longer Query

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

Models measured
345
Scores recorded
401
Sources
1
IQ weight
1.34

Score distribution

1,323.7
Mean
1,357.1
Median
137.1
Std dev
1,102.6
10th percentile
1,475.3
90th percentile
951.6
Min
951.61,549
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 345 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
201Trinity Large Preview· LMArena data1,355.3
202GPT-5 miniHigh· LMArena data1,355.2
203DeepSeek R1· LMArena data1,354.7
204Claude Sonnet 3.720250219· LMArena data1,354.7
205Qwen3 Next 80B A3Bthinking· LMArena data1,353.5
206Kimi K2 Preview0905· LMArena data1,352.5
207DeepSeek V30324· LMArena data1,352
208Qwen 3 235B A22B· LMArena data1,351.8
209Step-1o Turbo202506· LMArena data1,348.1
210GLM 4.7 Flash· LMArena data1,345.1
211Gemini 2.0 Flash001· LMArena data1,344.1
212O1 Preview· LMArena data1,343.7
213GPT-4.1 mini2025-04-14· LMArena data1,343.6
214DeepSeek V3· LMArena data1,342.8
215o3 MiniHigh· LMArena data1,342.8
216Hunyuan Large2025-02-10· LMArena data1,341.3
217Nova 2 Lite· LMArena data1,335.9
218Command A03-2025· LMArena data1,333.9
219Gemma 3 27B IT· LMArena data1,333.3
220nova-experimental-chat10-09· LMArena data1,332.1
221INTELLECT-3· LMArena data1,331.8
222MiniMax M2· LMArena data1,331.8
223Mercury 2· LMArena data1,330.3
224o3 Mini· LMArena data1,327.3
225Qwen3 32B· LMArena data1,327.1
226MiniMax M1· LMArena data1,326.4
227Kimi K2 Preview0711· LMArena data1,326
228Mistral Small 3.22506· LMArena data1,326
229Step3· LMArena data1,325.9
230Ling Flash 2.0· LMArena data1,324.3
231Qwen Plus0125· LMArena data1,324.2
232Granite 4.2 8B· LMArena data1,322.2
233Ring Flash 2.0· LMArena data1,320.4
234Gemini 2.0 Flash-Lite Preview02-05· LMArena data1,320.3
235O1 Mini· LMArena data1,320.2
236Hunyuan TurboS20250226· LMArena data1,320.1
237Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,319.5
238GPT OSS 120B· LMArena data1,318.9
239Trinity Large Thinking· LMArena data1,318.4
240Gemma 3 12B IT· LMArena data1,316.7
241GLM 4 Plus0111· LMArena data1,315.6
242o4 Mini2025-04-16· LMArena data1,314.4
243Llama 3.3 Nemotron Super 49B v1.5· LMArena data1,314.4
244Olmo 3.1 32B Instruct· LMArena data1,311.9
245GPT-5 nanoHigh· LMArena data1,311.7
246Qwen3 30B A3B· LMArena data1,311.6
247Claude 3.5 Sonnet20241022· LMArena data1,311
248Hunyuan Turbo0110· LMArena data1,308.7
249QwQ 32B· LMArena data1,307.8
250Gemini 1.5 Pro002· LMArena data1,307.6

Showing 201–250 of 401 results.

Data credits: LMArena, CC BY 4.0 (licence)