reasoning benchmark
FACTS Grounding Leaderboard
A benchmark evaluating language models' ability to generate factually accurate and well-grounded responses based on long-form input context, comprising 1,719 examples with documents up to 32k tokens requiring detailed responses that are fully grounded in provided documents
Updated Aug 17, 2026
Models13
Model coverage13
MetricScore
EvidenceC
FACTS Grounding Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelGO | Score87.8% | Percentile100.0% | Participants13 | EvidenceC | Evaluated |
| Rank02 | ModelGO | Score85.3% | Percentile91.7% | Participants13 | EvidenceC | Evaluated |
| Rank03 | ModelGO | Score84.1% | Percentile83.3% | Participants13 | EvidenceC | Evaluated |
| Rank04 | ModelGO | Score83.6% | Percentile75.0% | Participants13 | EvidenceC | Evaluated |
| Rank05 | ModelGO | Score83.6% | Percentile66.7% | Participants13 | EvidenceC | Evaluated |
| Rank06 | ModelGO | Score75.8% | Percentile58.3% | Participants13 | EvidenceC | Evaluated |
| Rank07 | ModelGO | Score74.9% | Percentile50.0% | Participants13 | EvidenceC | Evaluated |
| Rank08 | ModelGO | Score70.5% | Percentile41.7% | Participants13 | EvidenceC | Evaluated |
| Rank09 | ModelGO | Score70.1% | Percentile33.3% | Participants13 | EvidenceC | Evaluated |
| Rank10 | ModelGO | Score61.9% | Percentile25.0% | Participants13 | EvidenceC | Evaluated |
| Rank11 | ModelZA | Score58.6% | Percentile16.7% | Participants13 | EvidenceC | Evaluated |
| Rank12 | ModelGO | Score40.6% | Percentile8.3% | Participants13 | EvidenceC | Evaluated |
| Rank13 | ModelGO | Score36.4% | Percentile0.0% | Participants13 | EvidenceC | Evaluated |
FACTS Grounding Highlights
The leading models and scores on this benchmark.
FACTS Grounding Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for FACTS Grounding
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis facts grounding AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01GO
- Score
- 87.8%
- Price
- $1.3 input / $10 output per 1M tokens
- Speed
- Up to 85 tok/s via Google
Strengths
- Ranks #1 of 13 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures FACTS Grounding, not total model capability
- 02GOGemini 2.5 FlashGoogle
- Score
- 85.3%
- Price
- $0.30 input / $2.5 output per 1M tokens
- Speed
- Up to 85 tok/s via Google
Strengths
- Ranks #2 of 13 compared models
- 92th percentile on this benchmark
- C evidence result
Considerations
- This result measures FACTS Grounding, not total model capability
- 03GOGemini 2.5 Flash-LiteGoogle
- Score
- 84.1%
- Price
- $0.10 input / $0.40 output per 1M tokens
- Speed
- Up to 5.7 tok/s via Google
Strengths
- Ranks #3 of 13 compared models
- 83th percentile on this benchmark
- C evidence result
Considerations
- This result measures FACTS Grounding, not total model capability
- 04GOGemini 2.0 FlashGoogle
- Score
- 83.6%
- Speed
- Up to 183 tok/s via Google
Strengths
- Ranks #4 of 13 compared models
- 75th percentile on this benchmark
- C evidence result
Considerations
- This result measures FACTS Grounding, not total model capability
- 05GOGemini 2.0 Flash-LiteGoogle
- Score
- 83.6%
- Speed
- Up to 85 tok/s via Google
Strengths
- Ranks #5 of 13 compared models
- 67th percentile on this benchmark
- C evidence result
Considerations
- This result measures FACTS Grounding, not total model capability
Selection summary
Best AI Models for FACTS Grounding
Gemini 2.5 Pro Preview 06-05 currently leads FACTS Grounding with 87.8%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is FACTS Grounding?
What FACTS Grounding measures and how its scores work.
A benchmark evaluating language models' ability to generate factually accurate and well-grounded responses based on long-form input context, comprising 1,719 examples with documents up to 32k tokens requiring detailed responses that are fully grounded in provided documents
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of C.
- Family
- FACTS Grounding
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- facts-grounding|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about FACTS Grounding.
Which model scores highest on FACTS Grounding?
Gemini 2.5 Pro Preview 06-05 is currently ranked first with 87.8%.
What does FACTS Grounding measure?
A benchmark evaluating language models' ability to generate factually accurate and well-grounded responses based on long-form input context, comprising 1,719 examples with documents up to 32k tokens requiring detailed responses that are fully grounded in provided documents
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
13 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.