Skip to content

Benchmark

Arena ELO Text Factuality - Creative Writing

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

Models measured
142
Scores recorded
160
Sources
1
IQ weight
2.07

Score distribution

1,418.6
Mean
1,420.8
Median
43.1
Std dev
1,361.3
10th percentile
1,468.6
90th percentile
1,295.9
Min
1,295.91,508.7
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.07 for chat models, measured over 142 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101Doubao Seed 2.0 Pro· LMArena data1,408.6
102Grok 4.7xHigh· LMArena data1,408.5
103Nemotron 3 UltraNVFP4· LMArena data1,408.1
104Beam 1.0 501B A23B· LMArena data1,408
105MiMo V2 Omni· LMArena data1,405.9
106MiniMax M3· LMArena data1,405.1
107Grok 4.1 FastReasoning· LMArena data1,404.9
108Mistral Medium 3.12508· LMArena data1,404.3
109MiMo V2.5· LMArena data1,403
110Grok 4.3· LMArena data1,403
111Mistral Large 3· LMArena data1,401.8
112NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,400.7
113GPT 5.2· LMArena data1,400.4
114LongCat-Flash-Chat Exp2602· LMArena data1,400.4
115DeepSeek V3.2Thinking· LMArena data1,400.1
116Qwen3 Max Preview· LMArena data1,399.4
117Grok 4 FastReasoning· LMArena data1,397.8
118GPT-6 SolMax· LMArena data1,396.3
119Qwen 3.8 27B· LMArena data1,395.9
120Grok 40709· LMArena data1,395.8
121GPT-5.4 miniHigh· LMArena data1,395.2
122Qwen3 235B A22B Instruct 25072507· LMArena data1,394.8
123Claude Haiku 4.520251001· LMArena data1,394.1
124GPT 5.2High· LMArena data1,392.1
125Gemini 2.5 Flash Preview09-2025· LMArena data1,390.9
126MiMo V2 FlashNon-Thinking· LMArena data1,386.9
127Mistral Medium 3.5· LMArena data1,386.3
128Kimi K2 Thinking· LMArena data1,384.5
129Inkling 2 Small· LMArena data1,384.3
130Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,383.3
131MiniMax M2.7· LMArena data1,379.5
132MiniMax-M2.1 Preview· LMArena data1,378.5
133GPT-5 Chat· LMArena data1,378.3
134GPT-5.3 Chat· LMArena data1,376.5
135Qwen3.5 122B-A10B· LMArena data1,376
136Qwen3.5 27B· LMArena data1,375.1
137Step 3.5 Flash· LMArena data1,374.6
138Mistral Large 4· LMArena data1,373.6
139Muse Glimmer· LMArena data1,366.8
140nova-experimental-chat11-10· LMArena data1,366.6
141Qwen3.5 Flash· LMArena data1,365.8
142Claude Haiku 5.5High· LMArena data1,364.2
143Qwen3.5 35B A3B· LMArena data1,362.8
144Inkling Small· LMArena data1,361.3
145GPT 5High· LMArena data1,358.5
146Qwen3 Next 80B A3B Instruct· LMArena data1,354.1
147MiniMax M2.5· LMArena data1,353.9
148Solar Pro 4· LMArena data1,353.5
149MiMo V2 FlashThinking· LMArena data1,353.2
150GPT 5.4 NanoHigh· LMArena data1,353.1

Showing 101–150 of 160 results.

Data credits: LMArena, CC BY 4.0 (licence)