Skip to content

Benchmark

Arena ELO - Japanese

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

Models measured
245
Scores recorded
286
Sources
1
IQ weight
1.65

Score distribution

1,296.5
Mean
1,336.3
Median
150
Std dev
1,061.5
10th percentile
1,461.1
90th percentile
878.3
Min
878.31,557.4
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.65 for chat models, measured over 245 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101nova-experimental-chat11-10· LMArena data1,379.6
102Grok 4.6High· LMArena data1,379
103DeepSeek V3.1· LMArena data1,378.7
104DeepSeek V3.2Thinking· LMArena data1,378.5
105Grok 4.3· LMArena data1,378.5
106Claude Haiku 5.5High· LMArena data1,378.3
107Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,377.7
108LongCat-Flash-Chat Exp2602· LMArena data1,376.7
109ERNIE 5.00110· LMArena data1,376.2
110Grok 4.1Thinking· LMArena data1,376.1
111Claude Opus 4.120250805· LMArena data1,374.3
112MiMo V2.6 Flash· LMArena data1,372.6
113Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,372.4
114Qwen 3 235B A22BNon-Thinking· LMArena data1,372
115Qwen3.5 Flash· LMArena data1,369.1
116Qwen3.5 122B-A10B· LMArena data1,366.3
117GPT-5.4 miniHigh· LMArena data1,366.2
118Grok 4.7xHigh· LMArena data1,365.4
119GPT-6 LunaMax· LMArena data1,363.1
120MiMo V2 Pro· LMArena data1,360.7
121nova-experimental-chat10-20· LMArena data1,360.1
122DeepSeek V3.1Thinking· LMArena data1,359.6
123DeepSeek V3.2· LMArena data1,359.5
124Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,359.2
125Kimi K2 Thinking· LMArena data1,357.8
126GPT-6 SolMax· LMArena data1,357.7
127MiniMax M3· LMArena data1,353.8
128GPT-5.3 Chat· LMArena data1,353.7
129Grok 4 FastReasoning· LMArena data1,352.9
130Grok 4.1 FastReasoning· LMArena data1,352.4
131Step 3.5 Flash· LMArena data1,351.3
132Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,350
133Kimi K2 Preview0905· LMArena data1,349
134GLM 4.5 Air· LMArena data1,348.3
135o12024-12-17· LMArena data1,346.2
136Qwen3.5 27B· LMArena data1,346.2
137GPT 5.4 NanoHigh· LMArena data1,345.7
138Grok 3 Minibeta· LMArena data1,341.9
139Hunyuan TurboS20250416· LMArena data1,341.8
140GPT-5 miniHigh· LMArena data1,341.2
141Qwen3 30B A3B Instruct2507· LMArena data1,338
142Claude Haiku 4.520251001· LMArena data1,338
143Kimi K2 Preview0711· LMArena data1,336.3
144DeepSeek V30324· LMArena data1,333.2
145LongCat Flash Chat· LMArena data1,333.1
146Claude Opus 4Thinking 16K, 20250514· LMArena data1,331.7
147DeepSeek V3.2 Exp· LMArena data1,331.5
148GPT OSS 120B· LMArena data1,330.7
149Claude Opus 420250514· LMArena data1,329.8
150Solar Pro 4· LMArena data1,327

Showing 101–150 of 286 results.

Data credits: LMArena, CC BY 4.0 (licence)