Most Factually Accurate LLMs

The benchmarks behind the factuality ranking

Four public benchmarks decide this ranking. The heavier a benchmark's weight, the more it moves a model's score.

• 39%
  SimpleQA Verified

  Short factual questions, answered correctly

  Best on this testGPT-6 Astra75.6

• 21%
  Vectara HHEM hallucination rate

  How often its summaries add facts not in the source

  Best on this testFinix_s1_32b1.8%

• 20%
  AA-Omniscience Accuracy

  Answers hard knowledge questions correctly

  Best on this testClaude Fable 5.167.2

• 20%
  AA-Omniscience Non-hallucination

  Avoids making up answers it doesn't know

  Best on this testG9v3-3B88.3

How the factuality score is calculated

1. 1

   Rank on each benchmark

   Every model gets a percentile on each benchmark it has been measured on.

2. 2

   Steady the thin fields

   Where few models have taken a benchmark, that percentile is pulled toward the middle of the field.

3. 3

   Weigh and average

   The percentiles are averaged with the weights above into one score out of 100.

4. 4

   Qualify

   A model enters once it is measured on at least half the basket by weight, including one anchor benchmark.

357 models from 51 vendors have a factuality score, each measured on at least 2 of the 4 benchmarks. Scores come from AA-graded results, official model cards and third-party evaluations.

Questions about the factuality ranking