Skip to content

Benchmark

Arena ELO Text Factuality - Non English

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

Models measured
168
Scores recorded
192
Sources
1
IQ weight
2.36

Score distribution

1,422.4
Mean
1,430.7
Median
42.6
Std dev
1,361
10th percentile
1,469
90th percentile
1,245.6
Min
1,245.61,512.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.36 for chat models, measured over 168 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101Beam 1.0 501B A23B· LMArena data1,426.9
102DeepSeek V4 Flash· LMArena data1,426.8
103GLM 4.7· LMArena data1,426.5
104Gemini 3.1 Flash Lite· LMArena data1,425.8
105Gemini 2.5 Flash· LMArena data1,425.1
106Grok 4.7xHigh· LMArena data1,424.7
107Qwen 3.8 27B· LMArena data1,423
108MiMo V2 Pro· LMArena data1,422.5
109Qwen3.6 Plus· LMArena data1,422.2
110Grok 4.1Thinking· LMArena data1,421.7
111Claude Opus 4.120250805· LMArena data1,420.6
112DeepSeek V3.2 ExpThinking· LMArena data1,420
113DeepSeek-V4-Flash PreviewHigh· LMArena data1,419.7
114DeepSeek V3.2 Exp· LMArena data1,419.5
115Qwen3 235B A22B Instruct 25072507· LMArena data1,419.5
116DeepSeek V3.2· LMArena data1,418.9
117Mistral Large 3· LMArena data1,418.7
118Grok 4.1· LMArena data1,417.6
119MiniMax M3· LMArena data1,417.5
120MiMo V2.5· LMArena data1,417.2
121Nemotron 3 UltraNVFP4· LMArena data1,416.4
122Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,416.4
123MiMo V2 Omni· LMArena data1,414.5
124Inkling Small· LMArena data1,413.9
125Gemini 2.5 Flash Preview09-2025· LMArena data1,413.8
126Inkling 2 Small· LMArena data1,413.4
127Qwen3 Next 80B A3B Instruct· LMArena data1,413
128Doubao Seed 2.0 Pro· LMArena data1,412.8
129Qwen3 Coder· LMArena data1,412.8
130Kimi K2.5Instant· LMArena data1,412.4
131Grok 4.3· LMArena data1,411.9
132DeepSeek V3.2Thinking· LMArena data1,411.7
133Qwen3.5 122B-A10B· LMArena data1,411.6
134Mistral Medium 3.5· LMArena data1,411.5
135NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,411.3
136Claude Haiku 5.5High· LMArena data1,411
137Mistral Medium 3.12508· LMArena data1,410.1
138GPT-5 Chat· LMArena data1,410.1
139GPT 5High· LMArena data1,409.7
140Mistral Large 4· LMArena data1,407.7
141Grok 40709· LMArena data1,407.3
142GPT-5.3 Chat· LMArena data1,407
143nova-experimental-chat12-10· LMArena data1,405
144Kimi K2 Thinking· LMArena data1,404.2
145LongCat-Flash-Chat Exp2602· LMArena data1,403.1
146MiMo V2 FlashNon-Thinking· LMArena data1,402.9
147MiniMax M2.7· LMArena data1,402.1
148nova-experimental-chat11-10· LMArena data1,401.3
149Qwen3 VL 235B· LMArena data1,400.5
150GPT 5.4 NanoHigh· LMArena data1,400.4

Showing 101–150 of 192 results.

Data credits: LMArena, CC BY 4.0 (licence)