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
101GLM 4.5· LMArena data1,465.3
102Qwen3 235B A22B Instruct 25072507· LMArena data1,464.5
103nova-experimental-chat26-02-10· LMArena data1,463.7
104DeepSeek V4 Flash· LMArena data1,463.6
105MiniMax M3· LMArena data1,463.6
106LongCat-Flash-Chat Exp2602· LMArena data1,463.3
107Qwen3.5 122B-A10B· LMArena data1,463
108MiMo V2 FlashNon-Thinking· LMArena data1,462.6
109nova-experimental-chat11-10· LMArena data1,462.5
110DeepSeek V3.2 Exp· LMArena data1,462.2
111Inkling Small· LMArena data1,461.8
112DeepSeek V3.2 ExpThinking· LMArena data1,460.9
113Qwen3 VL 235B· LMArena data1,460.8
114Qwen3 Next 80B A3B Instruct· LMArena data1,460.6
115Gemini 3.1 Flash Lite· LMArena data1,460.2
116Claude Sonnet 4.520250929· LMArena data1,460
117GPT-5.2 Chat20260210· LMArena data1,459.5
118Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,458.4
119MiMo V2 Pro· LMArena data1,457.6
120Grok 4 FastNon-Reasoning· LMArena data1,457
121ChatGPT-4o Latest20250326· LMArena data1,456.5
122Qwen3.5 35B A3B· LMArena data1,456.3
123Kimi K2 Thinking· LMArena data1,455.5
124DeepSeek V3.2· LMArena data1,455.4
125GPT-5.5 Instant· LMArena data1,454.7
126NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,454
127Grok 4.1Thinking· LMArena data1,452.7
128Kimi K2.5Instant· LMArena data1,452.4
129DeepSeek V3.1· LMArena data1,452
130Gemini 2.5 Flash· LMArena data1,449.3
131Gemini 2.5 Flash Preview09-2025· LMArena data1,449.2
132Grok 3 Preview02-24· LMArena data1,448.2
133LongCat Flash Chat· LMArena data1,448.2
134DeepSeek V3.2Thinking· LMArena data1,447.9
135GPT 5.2High· LMArena data1,447.6
136Mistral Medium 3.12508· LMArena data1,447.2
137Qwen3.5 Flash· LMArena data1,447.1
138GPT 5.2· LMArena data1,446.1
139Step 3.5 Flash· LMArena data1,446.1
140Mistral Large 3· LMArena data1,445.8
141GPT-5.4 miniHigh· LMArena data1,445.2
142Mistral Medium 3.5· LMArena data1,444
143Qwen3 VL 235Bthinking· LMArena data1,442.6
144DeepSeek R10528· LMArena data1,441.7
145Grok 4.7xHigh· LMArena data1,440.5
146Grok 4.1 FastReasoning· LMArena data1,440.2
147MiniMax M2.7· LMArena data1,439.7
148nova-experimental-chat12-10· LMArena data1,438.1
149OpenAI o32025-04-16· LMArena data1,436.6
150Grok 4 FastReasoning· LMArena data1,435.1

Showing 101–150 of 396 results.

Data credits: LMArena, CC BY 4.0 (licence)