HealthBench

HealthBench grades a model's responses in realistic, multi-turn health conversations against physician-written rubrics, using another model as the judge.

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categorydomain
Subcategorymulti-turn health conversations, graded by a model against physician-written rubrics
Page statusactive
Metricrubric criteria met, normalized per conversation (mean score, clipped to [0,1])
Directionhigher_is_better
Unit%
Dataset size5000
Dataset licenceMIT for the evaluation code and released conversation/rubric data, per the openai/simple-evals GitHub repository. The paper separately asks that examples not be reposted online in plain text or images, to reduce the chance of leakage into future training corpora, and ships a canary string for filtering -- a usage request layered on top of, not a replacement for, the MIT code licence.
PublisherOpenAI

What it measures

HealthBench tests how a model responds across 5,000 realistic health conversations that stand in for a layperson, caregiver or clinician, produced through both synthetic generation and human adversarial testing so they are multi-turn (1 to 19 turns, averaging 2.6) and multilingual rather than single-shot trivia questions. Conversations are organised into seven themes -- emergency referrals, context seeking, global health, health data tasks, expertise-tailored communication, responding under uncertainty, and response depth -- each probing a different way a health-related answer can go right or wrong, from missing an emergency to using jargon with a layperson.

Task format

The model reads a multi-turn conversation and must produce the best possible final response to the user's last message. There is no fixed answer to match; every conversation instead carries its own physician-written rubric of scoring criteria that a grading model checks the response against.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub