Skip to content

Benchmark

FACTS Grounding

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

Models measured
13
Scores recorded
13
Sources
1
IQ weight
0.41

Score distribution

70.2%
Mean
74.9%
Median
16.1%
Std dev
40.6%
10th percentile
85.3%
90th percentile
36.4%
Min
36.4%87.8%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.41 for chat models, measured over 13 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini 2.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim87.8%
2Gemini 2.5 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim85.3%
3Gemini 2.5 Flash-Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim84.1%
4Gemini 2.0 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim83.6%
5Gemini 2.0 Flash Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim83.6%
6Gemma 3 12B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75.8%
7Gemma 3 27B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74.9%
8Gemini 3 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim70.5%
9Gemma 3 4B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim70.1%
10Gemini 3 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim61.9%
11GLM 5V Turbo· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim58.6%
12Gemini 3.1 Flash Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim40.6%
13Gemma 3 1b it· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim36.4%

Data credits: Data by LLM Stats