Skip to content

Benchmark

FACTS Benchmark Suite

Factuality benchmarks measure how often models get facts right. Scores are percentages. Higher is better.

Models measured
3
Scores recorded
3
Sources
1
IQ weight
No weight published

Score distribution

57.7%
Mean
61.9%
Median
12.6%
Std dev
40.6%
10th percentile
70.5%
90th percentile
40.6%
Min
40.6%70.5%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Gemini 3 Pro PreviewGemini 3 Pro Model Card · results as of November 2025· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.70.5%
2Gemini 3 FlashFACTS Benchmark Suite aggregate· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.61.9%
3Gemini 3.1 Flash LiteGemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.40.6%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.