HealthBench Concensus

AI benchmark results and model performance

Type: percentageHealth

86.97%

82.60% - 90.80%

82.60% - GPT OSS 20b

Scores Over Time

Individual benchmark scores plotted by date.

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
OpenAI	GPT OSS 20b	05 Aug 2025	84.90%	Low Reasoning Effort	Yes	Source
OpenAI	GPT OSS 120b	05 Aug 2025	90.80%	Medium Reasoning Effort	Yes	Source