FACTS Grounding Benchmark

Explore score history ↗

FACTS Grounding is an AI benchmark in the Benchmark Radar catalog.

Released: 2024-12-17
Openness: unknown
Importer: Benchmark Saturation

Source provenance

Original evidence

FACTS Grounding results and reported scores

Reported benchmark scores · Benchmark Saturation

Reported scores: 5

Model	Organization	Reported value	Reported	Evidence
Claude 3.5 Sonnet	—	82.2	—	evidence
Gemini 1.5 Flash	—	85.8	—	evidence
Gemini 1.5 Pro	—	82.7	—	evidence
Gemini 2.0 Flash Experimental	—	85.6	—	evidence
GPT-4o	—	79.8	—	evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.

Which sources cite FACTS Grounding?

FACTS Grounding