Skip to content

Benchmark

Arena ELO Text Factuality - Hard Prompts English

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

Models measured
156
Scores recorded
175
Sources
1
IQ weight
2.08

Score distribution

1,472.5
Mean
1,475.6
Median
36.4
Std dev
1,421.8
10th percentile
1,513.7
90th percentile
1,363.1
Min
1,363.11,545.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.08 for chat models, measured over 156 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 5.2· LMArena data1,468.8
102ERNIE 5.00110· LMArena data1,468.5
103GPT 5.2High· LMArena data1,468.4
104GLM 4.6· LMArena data1,468.2
105Qwen3 Max Preview· LMArena data1,467.8
106GLM 4.7· LMArena data1,467.8
107GPT-6 SolMax· LMArena data1,467.6
108Nemotron 3 UltraNVFP4· LMArena data1,466.4
109GPT-5.5 Instant· LMArena data1,465.8
110Claude Haiku 4.520251001· LMArena data1,465.7
111Mistral Large 4· LMArena data1,465.4
112ChatGPT-4o Latest20250326· LMArena data1,465.1
113NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,464.4
114Gemini 3 FlashMinimal· LMArena data1,464.4
115Grok 4.7xHigh· LMArena data1,464.2
116Mistral Large 3· LMArena data1,463.9
117MiMo V2 FlashNon-Thinking· LMArena data1,463.4
118DeepSeek V3.2 Exp· LMArena data1,463.3
119Inkling 2 Small· LMArena data1,462.8
120DeepSeek-V4-Flash PreviewHigh· LMArena data1,462.2
121Qwen3 235B A22B Instruct 25072507· LMArena data1,460.9
122Gemini 2.5 Flash· LMArena data1,459.3
123Solar Pro 4· LMArena data1,458.9
124Mistral Medium 3.12508· LMArena data1,457.5
125Claude Haiku 5.5High· LMArena data1,457
126Kimi K2 Thinking· LMArena data1,455.8
127Qwen3.5 27B· LMArena data1,455.6
128Inkling Small· LMArena data1,455.5
129ERNIE-5.0-Preview1203· LMArena data1,453.9
130Grok 4.1Thinking· LMArena data1,453.4
131Step 3.5 Flash· LMArena data1,452.3
132Grok 4 FastReasoning· LMArena data1,451.7
133Qwen3.5 122B-A10B· LMArena data1,450.9
134Grok 4.3· LMArena data1,449.9
135MiniMax-M2.1 Preview· LMArena data1,449.7
136Grok 4.1· LMArena data1,449.1
137nova-experimental-chat11-10· LMArena data1,448.8
138Doubao Seed 2.0 Pro· LMArena data1,447.8
139MiMo V2 FlashThinking· LMArena data1,446.8
140GPT 5.4 NanoHigh· LMArena data1,445.8
141GPT-5 Chat· LMArena data1,445.6
142Qwen3.5 Flash· LMArena data1,444.4
143Gemini 3.1 Flash Lite· LMArena data1,441.8
144Gemini 2.5 Flash Preview09-2025· LMArena data1,441.5
145GPT-5.3 Chat· LMArena data1,441.5
146nova-experimental-chat10-20· LMArena data1,440.6
147Qwen3.5 35B A3B· LMArena data1,439.8
148Grok 40709· LMArena data1,439.3
149Muse Glimmer· LMArena data1,434.1
150GPT 5High· LMArena data1,433.1

Showing 101–150 of 175 results.

Data credits: LMArena, CC BY 4.0 (licence)