Skip to content

Benchmark

Arena ELO Text Factuality

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

Models measured
170
Scores recorded
194
Sources
1
IQ weight
2.26

Score distribution

1,435.9
Mean
1,441.9
Median
39.3
Std dev
1,379.4
10th percentile
1,478.9
90th percentile
1,299.6
Min
1,299.61,522.2
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.26 for chat models, measured over 170 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101GPT 5.2High· LMArena data1,440.9
102Beam 1.0 501B A23B· LMArena data1,440.2
103ERNIE-5.0-Preview1022· LMArena data1,439.4
104GPT-6.2 AstraMax· LMArena data1,439.2
105DeepSeek V4 Flash· LMArena data1,439.1
106Grok 4.7xHigh· LMArena data1,438.8
107MiMo V2.5· LMArena data1,438.8
108Qwen3.6 Plus· LMArena data1,437.5
109DeepSeek V3.2· LMArena data1,436.8
110DeepSeek V3.2 Exp· LMArena data1,436.3
111Gemini 2.5 Flash· LMArena data1,436.2
112Qwen 3.8 27B· LMArena data1,436.2
113MiMo V2 Omni· LMArena data1,435
114Claude Opus 4.120250805· LMArena data1,434.9
115Grok 4.1Thinking· LMArena data1,434.6
116Mistral Large 3· LMArena data1,434.6
117DeepSeek-V4-Flash PreviewHigh· LMArena data1,433.9
118Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,433.2
119Gemini 3.1 Flash Lite· LMArena data1,433.1
120MiniMax M3· LMArena data1,433
121DeepSeek V3.2Thinking· LMArena data1,432.6
122Nemotron 3 UltraNVFP4· LMArena data1,432.3
123Grok 4.1· LMArena data1,431.7
124Qwen3 235B A22B Instruct 25072507· LMArena data1,430.9
125Mistral Medium 3.5· LMArena data1,429.8
126Kimi K2.5Instant· LMArena data1,429.1
127Mistral Medium 3.12508· LMArena data1,428.9
128LongCat-Flash-Chat Exp2602· LMArena data1,428.5
129Qwen3.5 122B-A10B· LMArena data1,427.8
130Grok 4.3· LMArena data1,426.3
131NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,426.2
132Inkling Small· LMArena data1,426.1
133Claude Haiku 5.5High· LMArena data1,424.4
134Gemini 2.5 Flash Preview09-2025· LMArena data1,424.2
135Inkling 2 Small· LMArena data1,424.1
136nova-experimental-chat12-10· LMArena data1,424
137Kimi K2 Thinking· LMArena data1,423
138MiMo V2 FlashNon-Thinking· LMArena data1,422.1
139GPT-5 Chat· LMArena data1,421.8
140MiniMax M2.7· LMArena data1,421.8
141Doubao Seed 2.0 Pro· LMArena data1,421
142Mistral Large 4· LMArena data1,421
143GPT 5High· LMArena data1,420.6
144Qwen3.5 27B· LMArena data1,419.2
145Qwen3 Next 80B A3B Instruct· LMArena data1,418.5
146Claude Haiku 4.520251001· LMArena data1,418.2
147Grok 40709· LMArena data1,418
148GPT-5.3 Chat· LMArena data1,417.6
149Grok 4 FastReasoning· LMArena data1,417
150GPT 5.4 NanoHigh· LMArena data1,416.1

Showing 101–150 of 194 results.

Data credits: LMArena, CC BY 4.0 (licence)