Skip to content

Benchmark

Arena ELO - Chinese

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

Models measured
340
Scores recorded
396
Sources
1
IQ weight
1.23

Score distribution

1,334.6
Mean
1,384.4
Median
172.2
Std dev
1,067.9
10th percentile
1,512.4
90th percentile
719
Min
7191,628.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.23 for chat models, measured over 340 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
201Qwen2.5-Max· LMArena data1,381.8
202Qwen 3 235B A22B· LMArena data1,380.4
203Step-1o Turbo202506· LMArena data1,379.6
204o3 MiniHigh· LMArena data1,378.6
205QwQ 32B· LMArena data1,378.5
206Grok 3 Minibeta· LMArena data1,377.7
207nova-experimental-chat26-01-10· LMArena data1,374
208Mistral Medium 32505· LMArena data1,373.6
209Gemini 2.0 Flash001· LMArena data1,372.9
210Trinity Large Thinking· LMArena data1,372.5
211MiniMax M2· LMArena data1,366.1
212Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,365.5
213Nova 2 Lite· LMArena data1,365.3
214Granite 4.2 8B· LMArena data1,363.9
215MiniMax M1· LMArena data1,360
216Hunyuan Turbo0110· LMArena data1,357.7
217Qwen3 32B· LMArena data1,356.9
218INTELLECT-3· LMArena data1,356.5
219Qwen3 Coder· LMArena data1,356.1
220GPT-5 nanoHigh· LMArena data1,356
221Trinity Large Preview· LMArena data1,354.4
222o4 Mini2025-04-16· LMArena data1,353.7
223Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,350.3
224Qwen Plus0125· LMArena data1,346.9
225Qwen3 30B A3B· LMArena data1,346.6
226Gemma 3 27B IT· LMArena data1,346.1
227Hunyuan Large2025-02-10· LMArena data1,345.9
228nova-experimental-chat10-09· LMArena data1,341
229Claude Sonnet 420250514· LMArena data1,340.9
230Mistral Small 3.22506· LMArena data1,340.1
231Gemini 2.0 Flash-Lite Preview02-05· LMArena data1,338.6
232DeepSeek V3· LMArena data1,338.3
233GLM 4.5V· LMArena data1,335.3
234Granite 4.1 8B· LMArena data1,334.5
235Gemini 1.5 Pro002· LMArena data1,331.3
236Step-2 16K Exp202412· LMArena data1,331.3
237GPT-4.1 mini2025-04-14· LMArena data1,328.6
238Command A03-2025· LMArena data1,327.3
239O1 Preview· LMArena data1,325.6
240Hunyuan TurboS20250226· LMArena data1,323.5
241Yi Lightning· LMArena data1,319.9
242o3 Mini· LMArena data1,319.4
243Hunyuan Standard2025-02-10· LMArena data1,319.2
244DeepSeek V2.51210· LMArena data1,317.6
245GPT OSS 20b· LMArena data1,314.9
246O1 Mini· LMArena data1,314.2
247Qwen2.5-Plus1127· LMArena data1,313.8
248Gemma 3n E4B IT· LMArena data1,308.9
249Olmo 3.1 32B Instruct· LMArena data1,305
250Athene V2 Chat· LMArena data1,301.9

Showing 201–250 of 396 results.

Data credits: LMArena, CC BY 4.0 (licence)