RAGTruth: provider-reported F1 (RAGTruth (provider reports))
We show this table for reference; we do not rank on it.
Corpus of retrieval-augmented responses annotated word by word for unsupported or contradicted content.
Benchmark score on RAGTruth (provider reports) — October 29, 2026
We compile the RAGTruth (provider reports) rows from provider self-reports. The table contains 2 models. Nemotron 3.5 Bolt E4B has the highest published score at 77.88%, but the available coverage and evaluation setups do not establish a market leader. We do not use these results to rank models overall.
Nemotron 3.5 Bolt E4B
NVIDIA
Nemotron 3.5 Bolt E4B BF16
NVIDIA
2 modelsKnowledgeRefreshingDisplay onlyUpdated October 29, 2026
Benchmark score table (2 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Nemotron 3.5 Bolt E4BNVIDIA | 77.88% | Not reported | Open |
| 2 | Nemotron 3.5 Bolt E4B BF16NVIDIA | 77.59% | Not reported | Open |
How to read this leaderboard
Operator receipt: 2 sourced rows are currently displayable on this page; the highest published score among these 2 models is Nemotron 3.5 Bolt E4B at 77.88%, which does not establish a market leader.
Honest limit: The task setup behind the F1 is not stated. Only NVIDIA-run results are stored so far, from one harness. Display-only and excluded from weighted ranking.
About RAGTruth (provider reports)
Year
2024
Tasks
Retrieval-augmented question answering, summarization and data-to-text
Format
F1
Difficulty
Avoiding unsupported claims
NVIDIA reports it for Nemotron 3.5 Bolt E4B (October 23, 2026), run in its NeMo Gym harness, as F1 in reasoning mode, and describes the task as avoiding unsupported claims. NVIDIA does not say whether the model wrote the responses or labeled them. The RAGTruth paper board keeps its generator and detector profiles separately.
Freshness and provenance
Version
RAGTruth (provider reports) 2024
Refresh cadence
Annual
Staleness state
Refreshing
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does RAGTruth (provider reports) measure?
Corpus of retrieval-augmented responses annotated word by word for unsupported or contradicted content.
Which model scores highest on RAGTruth (provider reports)?
Nemotron 3.5 Bolt E4B by NVIDIA currently leads with a score of 77.88% on RAGTruth (provider reports).
How many models are evaluated on RAGTruth (provider reports)?
2 AI models have published results on RAGTruth (provider reports) in the BenchLM catalog.
Compare top models on RAGTruth (provider reports)
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 6,100+ subscribers.
One email each week. Unsubscribe anytime.