Skip to content

Benchmark

Arena ELO Text Factuality - Longer Query

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

Models measured
160
Scores recorded
179
Sources
1
IQ weight
2.18

Score distribution

1,450.6
Mean
1,455.4
Median
42.5
Std dev
1,387.5
10th percentile
1,494.5
90th percentile
1,307.2
Min
1,307.21,538.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.18 for chat models, measured over 160 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101Kimi K2.5Instant· LMArena data1,449.3
102GPT 5.2· LMArena data1,448.5
103Nemotron 3 UltraNVFP4· LMArena data1,448
104Grok 4.20 Beta 1· LMArena data1,448
105Grok 4.7xHigh· LMArena data1,447.7
106Qwen3 235B A22B Instruct 25072507· LMArena data1,447.7
107DeepSeek V3.2Thinking· LMArena data1,447.6
108Gemma 4 26B A4B· LMArena data1,447.1
109GPT-5.4 miniHigh· LMArena data1,447.1
110GLM 4.7· LMArena data1,447.1
111DeepSeek-V4-Flash PreviewHigh· LMArena data1,446.6
112ChatGPT-4o Latest20250326· LMArena data1,445.6
113GPT-5.2 Chat20260210· LMArena data1,444.6
114DeepSeek V3.2 Exp· LMArena data1,444.1
115Gemini 2.5 Flash· LMArena data1,443.9
116MiniMax M2.7· LMArena data1,443.7
117Mistral Large 3· LMArena data1,441.1
118GPT 5.2High· LMArena data1,441
119LongCat-Flash-Chat Exp2602· LMArena data1,440.9
120GLM 4.6· LMArena data1,440.8
121MiniMax-M2.1 Preview· LMArena data1,439.6
122Mistral Medium 3.5· LMArena data1,439.3
123ERNIE 5.00110· LMArena data1,437.8
124ERNIE-5.0-Preview1203· LMArena data1,436.7
125Qwen3.5 27B· LMArena data1,435
126Grok 4.3· LMArena data1,434.6
127Inkling Small· LMArena data1,433.2
128Kimi K2 Thinking· LMArena data1,432.9
129MiMo V2 FlashNon-Thinking· LMArena data1,432.8
130Grok 4 FastReasoning· LMArena data1,432.6
131Doubao Seed 2.0 Pro· LMArena data1,432.1
132Mistral Medium 3.12508· LMArena data1,430
133Grok 4.1· LMArena data1,429.2
134NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,428.4
135Qwen3.5 122B-A10B· LMArena data1,428.2
136GPT-5.3 Chat· LMArena data1,427.9
137Gemini 2.5 Flash Preview09-2025· LMArena data1,427.9
138Inkling 2 Small· LMArena data1,427.9
139Mistral Large 4· LMArena data1,427.6
140Gemini 3.1 Flash Lite· LMArena data1,426.5
141Grok 4.1Thinking· LMArena data1,426.3
142GPT-5 Chat· LMArena data1,426
143Step 3.5 Flash· LMArena data1,424.6
144Claude Haiku 5.5High· LMArena data1,423.7
145Qwen3.5 Flash· LMArena data1,422.2
146Grok 40709· LMArena data1,418.5
147Solar Pro 4· LMArena data1,417.5
148GPT 5.4 NanoHigh· LMArena data1,416
149Qwen3.5 35B A3B· LMArena data1,415.4
150MiMo V2 FlashThinking· LMArena data1,414.7

Showing 101–150 of 179 results.

Data credits: LMArena, CC BY 4.0 (licence)