Skip to content

Benchmark

Arena ELO Text Factuality - Industry Writing and Literature and Language

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

Models measured
154
Scores recorded
173
Sources
1
IQ weight
2.19

Score distribution

1,424.1
Mean
1,427.8
Median
44.3
Std dev
1,368.5
10th percentile
1,472.9
90th percentile
1,271.6
Min
1,271.61,520.8
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.19 for chat models, measured over 154 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101DeepSeek V3.2· LMArena data1,420.2
102GLM 4.6· LMArena data1,419.9
103MiMo V2.5· LMArena data1,419.7
104GPT-5.4 miniHigh· LMArena data1,419.2
105Beam 1.0 501B A23B· LMArena data1,418.6
106Grok 4.1Thinking· LMArena data1,418.2
107Gemini 2.5 Flash· LMArena data1,417.8
108Grok 4.1· LMArena data1,416.7
109Nemotron 3 UltraNVFP4· LMArena data1,415.1
110DeepSeek V3.2Thinking· LMArena data1,414.6
111GPT-6 SolMax· LMArena data1,413.9
112GPT 5.2High· LMArena data1,412.5
113Doubao Seed 2.0 Pro· LMArena data1,411.8
114MiMo V2 Omni· LMArena data1,411.5
115Mistral Large 3· LMArena data1,411.2
116Qwen 3.8 27B· LMArena data1,411.1
117Grok 4.3· LMArena data1,410.7
118Grok 40709· LMArena data1,410.7
119Claude Haiku 4.520251001· LMArena data1,409.1
120Qwen3 235B A22B Instruct 25072507· LMArena data1,407.7
121Mistral Medium 3.12508· LMArena data1,406.9
122Gemini 2.5 Flash Preview09-2025· LMArena data1,406.6
123Qwen3 Max Preview· LMArena data1,406.1
124Kimi K2.5Instant· LMArena data1,405.9
125NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,404.9
126Inkling 2 Small· LMArena data1,404.3
127GPT-5 Chat· LMArena data1,403.7
128LongCat-Flash-Chat Exp2602· LMArena data1,401.5
129Grok 4 FastReasoning· LMArena data1,400.9
130GPT 4.12025-04-14· LMArena data1,399.8
131Mistral Medium 3.5· LMArena data1,399.6
132Grok 4.1 FastReasoning· LMArena data1,398.1
133Qwen3.5 27B· LMArena data1,397.1
134GPT-5.3 Chat· LMArena data1,397
135Qwen3.5 122B-A10B· LMArena data1,396.9
136Kimi K2 Thinking· LMArena data1,396.6
137MiniMax M2.7· LMArena data1,395
138MiMo V2 FlashNon-Thinking· LMArena data1,394.3
139Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,394.2
140Step 3.5 Flash· LMArena data1,393
141Claude Haiku 5.5High· LMArena data1,391.7
142Qwen3.5 Flash· LMArena data1,389.6
143MiniMax-M2.1 Preview· LMArena data1,388.2
144GPT 5High· LMArena data1,387.6
145nova-experimental-chat11-10· LMArena data1,386.6
146Mistral Large 4· LMArena data1,385.9
147Inkling Small· LMArena data1,385.6
148Muse Glimmer· LMArena data1,383.7
149Qwen3.5 35B A3B· LMArena data1,383.1
150OpenAI o32025-04-16· LMArena data1,380.4

Showing 101–150 of 173 results.

Data credits: LMArena, CC BY 4.0 (licence)