# FACTS Grounding V1 > The archived FACTS Grounding V1 protocol tests factual responses grounded in supplied documents. Its original judge setup and results remain separate from Grounding V2. Canonical page: https://benchlm.ai/benchmarks/facts-grounding-v1 - Category: [Reasoning](/reasoning) - Last updated: Archived version 1 Evaluation results published by [Google DeepMind, Google Research, and Kaggle](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1). undefined The exact version 1 page declares Apache 2.0. We retain aggregate measurements, attribution, and source links. No tasks, questions, answers, documents, or other underlying dataset contents are included. [Original article terms](https://www.apache.org/licenses/LICENSE-2.0). FACTS Grounding V1 tests responses to requests accompanied by documents of up to 32,000 tokens. Its 1,719 examples are split into 860 public and 859 private examples. The original protocol uses an ensemble of three model judges and filters responses for eligibility before assessing factual grounding. The version 1 download supplies overall, public, and private scores. Version 2 changes the evaluation setup, so the archived scores remain separate. This board measures support from the supplied document; it does not establish factual accuracy outside that document. The download was acquired on the recorded retrieval date. Per-result evaluation timestamps are publisher dates, not evidence of when this leaderboard snapshot was first available. These archived version 1 protocols remain separate from the current version 2 boards. The complete official download includes configurations hidden by default source filters. Source identities are preserved, and these measurements do not enter overall or category rankings. These archived version 1 protocols remain separate from the current version 2 boards. The complete official download includes configurations hidden by default source filters. Source identities are preserved, and these measurements do not enter overall or category rankings. - [Official archived Kaggle leaderboard](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) - [Official version 1 leaderboard JSON](https://www.kaggle.com/api/v1/benchmarks/google/facts-grounding/versions/1/leaderboard) - [Evaluation method](https://arxiv.org/abs/2501.03200) - [Benchmark-page Apache 2.0 declaration](https://www.apache.org/licenses/LICENSE-2.0) ## About FACTS Grounding V1 - Year: 2025 - Tasks: 1,719 examples: 860 public and 859 private - Format: Document-grounded answering with an ensemble of three judges - Paper: [FACTS Grounding V1 methodology](https://arxiv.org/abs/2501.03200) FACTS Grounding V1 tests responses to requests accompanied by documents of up to 32,000 tokens. Its 1,719 examples are split into 860 public and 859 private examples. The original protocol uses an ensemble of three model judges and filters responses for eligibility before assessing factual grounding. The version 1 download supplies overall, public, and private scores. Version 2 changes the evaluation setup, so the archived scores remain separate. This board measures support from the supplied document; it does not establish factual accuracy outside that document. These archived version 1 protocols remain separate from the current version 2 boards. The complete official download includes configurations hidden by default source filters. Source identities are preserved, and these measurements do not enter overall or category rankings. FACTS Grounding V1 is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula. ## Leaderboard (35 source rows) | Rank | Model | Configuration | Creator | Score | |------|-------|---------------|---------|-------| | 1 | [Llama 3 - Grounded Language Model](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-08-20 | — | 89.4% | | 2 | [Gemini 2.5 Pro](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-06-06 | — | 87.8% | | 3 | [GPT-5](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-08-11 | — | 87.4% | | 4 | [Gemini 2.5 Flash](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-07-14 | — | 86.4% | | 5 | [Gemini 2.0 Flash](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 85.6% | | 6 | [Claude 3.5 Sonnet](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 83.3% | | 7 | [Gemini 2.0 Flash Lite](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-06-20 | — | 83.2% | | 8 | [DeepSeek-R1](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-06-04 | — | 82.4% | | 9 | [Gemini 1.5 Flash](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-16 | — | 82.4% | | 10 | [Gemini 1.5 Pro](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 81.2% | | 11 | [Gemma 3 27B](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 80.6% | | 12 | [Gemma 3 12B](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 80.5% | | 13 | [Gemini 1.5 Flash 8B](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-16 | — | 79.7% | | 14 | [Claude Sonnet 4](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-23 | — | 79.1% | | 15 | [Claude 3.7 Sonnet](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 78.8% | | 16 | [o1](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 78.8% | | 17 | [Claude 3.5 Haiku](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 78.1% | | 18 | [Claude Opus 4](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-23 | — | 77.7% | | 19 | [o3 mini](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 77.6% | | 20 | [Grok 2](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 76.2% | | 21 | [GPT-4o](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 75.9% | | 22 | [Grok 3](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-29 | — | 75.7% | | 23 | [Grok 3 Mini](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-26 | — | 75.0% | | 24 | [GPT-4.1](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 74.5% | | 25 | [Mistral Large 2](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 72.6% | | 26 | [Mixtral 8x22B](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 70.1% | | 27 | [o3](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 69.6% | | 28 | [GPT-4o mini](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 68.4% | | 29 | [Gemma 3 4B](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 64.8% | | 30 | [DeepSeek-V3](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 63.9% | | 31 | [o1 mini](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 62.1% | | 32 | [o4 mini](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 62.1% | | 33 | [Codestral](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 59.7% | | 34 | [Ministral 8B](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 56.1% | | 35 | [Ministral 3B](https://www.kaggle.com/benchmarks/google/facts-grounding/versions/1) | Archived Kaggle benchmark v1; Task revision 1; Evaluated 2025-05-08 | — | 52.2% | ## FAQ ### What does FACTS Grounding V1 measure? The archived FACTS Grounding V1 protocol tests factual responses grounded in supplied documents. Its original judge setup and results remain separate from Grounding V2. ### Which source row has the highest reported FACTS Grounding V1 value? Llama 3 - Grounded Language Model has the highest reported FACTS Grounding V1 value at 89.4%. These archived version 1 protocols remain separate from the current version 2 boards. The complete official download includes configurations hidden by default source filters. Source identities are preserved, and these measurements do not enter overall or category rankings. ### How many models are evaluated on FACTS Grounding V1? The Archived version 1 contains 35 AI models. ### Does FACTS Grounding V1 affect BenchLM's overall score? Not directly. FACTS Grounding V1 is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.