GPQA: A Graduate-Level Google-Proof Q&A Benchmark

Explore score history ↗

GPQA: A Graduate-Level Google-Proof Q&A Benchmark is an AI benchmark in the Benchmark Radar catalog.

Released: 2023-11-21
Openness: unknown
Importer: Benchmark Saturation

Source provenance

Original evidence

GPQA: A Graduate-Level Google-Proof Q&A Benchmark results and reported scores

Reported benchmark scores · Benchmark Saturation

Reported scores: 5

Model	Organization	Reported value	Reported	Evidence
Claude-4.5 Sonnet	—	83.4	—	evidence
Gemini-2.5 Pro	—	84.4	—	evidence
GLM 4.6	—	82.9	—	evidence
GPT-5	—	85.4	—	evidence
Grok4	—	87.7	—	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 GPQA: A Graduate-Level Google-Proof Q&A Benchmark?