FACTS Benchmark Suite
General QA · 2025-12-11
The FACTS Benchmark Suite is Google DeepMind's aggregate factuality score, the unweighted average of four component leaderboards: FACTS Grounding, FACTS Parametric, FACTS Search, and FACTS Multimodal. It is a later and broader construct than FACTS Grounding alone, which it contains as one component, and the two are not interchangeable.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 3 Pro Preview | 70.5 |
| Gemini 3.1 Pro Preview | 67.3 |
| GPT-5.5 | 66.6 |
| Gemini 3.6 Flash | 65.3 |
| Gemini 3 Flash Preview | 63.4 |
| Gemini 3.5 Flash | 62.3 |
| GPT-5.2 | 61.4 |
| GPT-5 (2025-08-07) | 56.5 |
| Opus 4.8 | 53.5 |
| Gemini 3.5 Flash-Lite | 53.5 |
| GPT-5.6 Sol | 52.4 |
| GPT-5.6 Terra | 52.0 |
| Gemini 3.1 Flash-Lite Preview | 51.4 |
| GPT-5.1 | 50.8 |
| Grok 4.5 | 50.2 |
| GPT-5.2 (2025-12-11) | 48.5 |
| Grok 4 | 46.3 |
| GPT-4.1 (2025-04-14) | 45.8 |
| Opus 4.7 | 45.4 |
| Gemma 4 31B IT | 45.3 |
| O3 (2025-04-16) | 44.4 |
| Opus 4.5 | 43.9 |
| Gemma 4 26B A4B IT | 42.7 |
| Grok 4.1 Fast | 42.1 |
| Opus 4.6 | 41.4 |
| Sonnet 4.5 | 40.6 |
| Opus 4.1 | 40.4 |
| GPT-5 Mini (2025-08-07) | 38.6 |
| Sonnet 4.6 | 37.0 |
| Sonnet 4 | 35.0 |
| GPT-5.6 Luna | 34.8 |
| GPT-5 Mini | 33.7 |
| Gemma 3 27B IT | 29.1 |
| O4 Mini (2025-04-16) | 28.1 |
| Grok 4 Fast | 25.5 |
| Haiku 4.5 | 18.6 |
| Gemini 2.5 Flash-Lite | 17.9 |