HellaSwag Benchmark

Explore score history ↗

HellaSwag is an AI benchmark in the Benchmark Radar catalog.

Released: 2019-05-20
Openness: unknown
Importer: Benchmark Saturation

Source provenance

Original evidence

HellaSwag results and reported scores

Reported benchmark scores · Benchmark Saturation

Reported scores: 5

Model	Organization	Reported value	Reported	Evidence
Claude 3 Opus	—	95.4	—	evidence
Claude 3 Sonnet	—	89.0	—	evidence
Gemini 1.5 Pro	—	93.3	—	evidence
GPT-4	—	95.3	—	evidence
Kimi-k2-base	—	94.6	—	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 HellaSwag?

HellaSwag