Natural Questions Benchmark

Explore score history ↗

Natural Questions is an AI benchmark in the Benchmark Radar catalog.

Released: 2019-07-01
Openness: unknown
Importer: Benchmark Saturation

Source provenance

Original evidence

Natural Questions results and reported scores

Reported benchmark scores · Benchmark Saturation

Reported scores: 5

Model	Organization	Reported value	Reported	Evidence
Amazon Nova Lite	—	81.5	—	evidence
Amazon Nova Pro	—	82.9	—	evidence
GPT-4 Turbo (2024-04-09)	—	79.5	—	evidence
GPT-4o (2024-05-13)	—	80.3	—	evidence
PaLM-2 (Bison)	—	81.3	—	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 Natural Questions?

Natural Questions