# FACTS Benchmark Suite

> The official FACTS suite combines document grounding, visual factuality, parametric recall, and search-assisted answering. The published average stays separate from its component scores.

Canonical page: https://benchlm.ai/benchmarks/facts

- Category: [Knowledge](/knowledge)
- Last updated: October 13, 2026

Evaluation results published by [Google DeepMind, Google Research, and Kaggle](https://www.kaggle.com/benchmarks/google/facts). undefined

Kaggle labels this benchmark Apache 2.0. We retain aggregate measurements with attribution and source links. That declaration does not establish rights to third-party documents, videos, or other underlying task material; no task contents are included. [Original article terms](https://www.apache.org/licenses/LICENSE-2.0).

The FACTS suite brings four factuality tests together: Grounding v2, Multimodal v1, Parametric v1, and Search v2 in this source snapshot. We show Kaggle's published Average without recalculating it or borrowing scores from provider reports.

Some rows lack a component result even though the table labels the average as four dimensions. The retained measurements preserve those missing cells. A published average does not establish that a model completed all four tests, and it should be read alongside its coverage.

The complete official download includes models hidden by the source page's default filters. Source names and task versions stay separate from canonical model profiles; these results do not enter overall or category rankings.

These are published source configurations. Compare the task version and disclosed settings before treating a score difference as a model difference.

- [Official Kaggle leaderboard](https://www.kaggle.com/benchmarks/google/facts)
- [Official leaderboard JSON](https://www.kaggle.com/api/v1/benchmarks/google/facts/leaderboard)
- [Evaluation method](https://arxiv.org/abs/2512.10791)
- [Benchmark-page Apache 2.0 declaration](https://www.apache.org/licenses/LICENSE-2.0)
- [FACTS suite introduction](https://deepmind.google/blog/facts-benchmark-suite-systematically-evaluating-the-factuality-of-large-language-models/)

## About FACTS Benchmark Suite

- Year: 2025
- Tasks: Four benchmark dimensions; public and held-out examples
- Format: Published percentage average
- Paper: [FACTS Benchmark Suite methodology](https://arxiv.org/abs/2512.10791)

The FACTS suite brings four factuality tests together: Grounding v2, Multimodal v1, Parametric v1, and Search v2 in this source snapshot. We show Kaggle's published Average without recalculating it or borrowing scores from provider reports. Some rows lack a component result even though the table labels the average as four dimensions. The retained measurements preserve those missing cells. A published average does not establish that a model completed all four tests, and it should be read alongside its coverage. The complete official download includes models hidden by the source page's default filters. Source names and task versions stay separate from canonical model profiles; these results do not enter overall or category rankings.

FACTS Benchmark Suite is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (40 source rows)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [GPT-6 Astra](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-09-15 | — | 75.7% |
| 2 | [GPT-6.1 Sol](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-10-13 | — | 75.3% |
| 3 | [GPT-6 Sol](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-09-29 | — | 72.5% |
| 4 | [Gemini 3.7 Flash](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-08-27 | — | 71.2% |
| 5 | [Gemini 3.8 Flash](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-09-04 | — | 69.4% |
| 6 | [Gemini 3.1 Pro Preview](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-07-25 | — | 67.4% |
| 7 | [GPT-5.5](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluation date unreported | — | 66.7% |
| 8 | [Gemini 3.6 Flash](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-07-23 | — | 65.3% |
| 9 | [Claude Opus 5.5](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-09-25 | — | 65.3% |
| 10 | [Gemini 3 Flash Preview](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-07-18 | — | 63.4% |
| 11 | [GPT-6 Luna](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-09-29 | — | 63.3% |
| 12 | [Gemini 3.5 Flash](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-07-23 | — | 63.1% |
| 13 | [Gemini 2.5 Pro](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 61.5% |
| 14 | [GPT-5](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 56.5% |
| 15 | [Claude Opus 4.8](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-07-24 | — | 53.5% |
| 16 | [Gemini 3.5 Flash-Lite](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-07-23 | — | 53.5% |
| 17 | [GPT-5.6 Sol](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2026-07-24 | — | 52.4% |
| 18 | [GPT-5.6 Terra](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-07-24 | — | 52.0% |
| 19 | [Gemini 3.1 Flash-Lite Preview](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluation date unreported | — | 51.4% |
| 20 | [Grok 4.5](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-07-28 | — | 50.2% |
| 21 | [GPT-5.2](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluation date unreported | — | 48.5% |
| 22 | [Gemini 2.5 Flash](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 47.2% |
| 23 | [Grok 4](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 46.3% |
| 24 | [GPT-4.1](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 45.8% |
| 25 | [Claude Opus 4.7](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluation date unreported | — | 45.5% |
| 26 | [Gemma 4 31B](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-07-27 | — | 45.3% |
| 27 | [GPT-5.1](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 45.0% |
| 28 | [o3](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 44.4% |
| 29 | [Claude Opus 4.5](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 43.9% |
| 30 | [Gemma 4 26B A4B](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-07-28 | — | 42.7% |
| 31 | [Claude Opus 4.6](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluation date unreported | — | 41.5% |
| 32 | [Claude Sonnet 4.5 (thinking)](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 40.6% |
| 33 | [Claude Opus 4.1](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 40.4% |
| 34 | [GPT-5 mini](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 38.6% |
| 35 | [Claude Sonnet 4.6](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluation date unreported | — | 37.0% |
| 36 | [GPT-5.6 Luna](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 4/4 component scores; Evaluated 2026-08-12 | — | 35.0% |
| 37 | [Claude Sonnet 4](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 35.0% |
| 38 | [Gemma 3 27B](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 29.1% |
| 39 | [o4 mini](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 28.1% |
| 40 | [grok-4-fast-reasoning](https://www.kaggle.com/benchmarks/google/facts) | Kaggle benchmark v1; task revision 1; 3/4 component scores; Evaluated 2025-12-07 | — | 25.5% |

## FAQ

### What does FACTS Benchmark Suite measure?

The official FACTS suite combines document grounding, visual factuality, parametric recall, and search-assisted answering. The published average stays separate from its component scores.

### Which source row has the highest reported FACTS Benchmark Suite value?

GPT-6 Astra has the highest reported FACTS Benchmark Suite value at 75.7%. These are published source configurations. Compare the task version and disclosed settings before treating a score difference as a model difference.

### How many models are evaluated on FACTS Benchmark Suite?

The October 13, 2026 contains 40 AI models.

### Does FACTS Benchmark Suite affect BenchLM's overall score?

Not directly. FACTS Benchmark Suite is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.