Skip to content

Benchmark

Arena ELO Text Factuality - Exclude Ties

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

Models measured
169
Scores recorded
193
Sources
1
IQ weight
2.31

Score distribution

1,431.5
Mean
1,440
Median
52.6
Std dev
1,353.3
10th percentile
1,488
90th percentile
1,239.9
Min
1,239.91,545.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.31 for chat models, measured over 169 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 data1,437.3
102ERNIE-5.0-Preview1022· LMArena data1,436.4
103Qwen3.6 Plus· LMArena data1,435.8
104Qwen 3.8 27B· LMArena data1,435.3
105GPT-6.2 AstraMax· LMArena data1,434.3
106MiMo V2.5· LMArena data1,434.1
107Grok 4.1Thinking· LMArena data1,433.9
108GPT 5.2· LMArena data1,433.6
109DeepSeek V3.2· LMArena data1,433.4
110GPT 5.2High· LMArena data1,433.1
111GPT-6 SolMax· LMArena data1,432.8
112Mistral Large 3· LMArena data1,432.7
113Nemotron 3 UltraNVFP4· LMArena data1,430.8
114Grok 4.1· LMArena data1,430.8
115MiMo V2 Omni· LMArena data1,430.4
116MiniMax M3· LMArena data1,430.1
117DeepSeek-V4-Flash PreviewHigh· LMArena data1,430
118Grok 4.7xHigh· LMArena data1,429.9
119Gemini 2.5 Flash· LMArena data1,429.9
120Claude Opus 4.120250805· LMArena data1,429.1
121Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,426.6
122Mistral Medium 3.5· LMArena data1,426.3
123Gemini 3.1 Flash Lite· LMArena data1,426.2
124LongCat-Flash-Chat Exp2602· LMArena data1,426
125DeepSeek V3.2Thinking· LMArena data1,425.8
126Mistral Medium 3.12508· LMArena data1,425.7
127Qwen3 235B A22B Instruct 25072507· LMArena data1,425.4
128Kimi K2.5Instant· LMArena data1,425.3
129Doubao Seed 2.0 Pro· LMArena data1,423
130Qwen3.5 122B-A10B· LMArena data1,420.7
131NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,420.5
132Inkling Small· LMArena data1,420.3
133nova-experimental-chat12-10· LMArena data1,418.4
134Grok 4.3· LMArena data1,417.6
135Gemini 2.5 Flash Preview09-2025· LMArena data1,416.2
136Kimi K2 Thinking· LMArena data1,416.1
137Claude Haiku 5.5High· LMArena data1,415.7
138MiMo V2 FlashNon-Thinking· LMArena data1,415.3
139MiniMax M2.7· LMArena data1,415.2
140Mistral Large 4· LMArena data1,415.1
141Inkling 2 Small· LMArena data1,413.8
142GPT-5 Chat· LMArena data1,411.3
143Qwen3 Next 80B A3B Instruct· LMArena data1,410.4
144Qwen3.5 27B· LMArena data1,409.5
145nova-experimental-chat11-10· LMArena data1,408.2
146Claude Haiku 4.520251001· LMArena data1,408.1
147GPT 5High· LMArena data1,407.6
148Grok 40709· LMArena data1,407.5
149Step 3.5 Flash· LMArena data1,406.5
150Qwen3 VL 235B· LMArena data1,405.8

Showing 101–150 of 193 results.

Data credits: LMArena, CC BY 4.0 (licence)