Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Knowledge benchmark
Every model the catalog carries a published HealthBench Professional value for, ranked by that value.
An open benchmark for clinician-facing model responses across care consult, writing and documentation, and medical research tasks. 21 models in the ModelChoice catalog have a published value.
Each point is a model with a published HealthBench Professional value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
21 models with a published HealthBench Professional value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Rubric-graded open-ended responses |
|---|---|---|---|
| 1 | Anthropic | 69.2 | |
| 2 | OpenAI | 66.3 | |
| 3 | OpenAI | 65.9 |
What the benchmark measures and how its value is published, from the source’s own record.
HealthBench Professional is a knowledge benchmark published by HealthBench Professional: Evaluating Large Language Models on Real Clinician Chats. An open benchmark for clinician-facing model responses across care consult, writing and documentation, and medical research tasks. Its evaluation set is described by the source as clinician chat tasks. The source rates its difficulty as professional clinical workflows.
Each model's result is published by HealthBench Professional: Evaluating Large Language Models on Real Clinician Chats as rubric-graded open-ended responses. ModelChoice prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 21 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
| 4 | Anthropic | 65.6 |
| 5 | InclusionAI | 65.3 |
| 6 | OpenAI | 64.7 |
| 7 | OpenAI | 64.2 |
| 8 | OpenAI | 62.7 |
| 9 | Meta | 62.3 |
| 10 | OpenAI | 60.8 |
| 10 | OpenAI | 60.8 |
| 12 | OpenAI | 60.5 |
| 12 | Meta | 60.5 |
| 14 | Alibaba | 60.0 |
| 15 | Anthropic | 59.8 |
| 16 | Meta | 59.3 |
| 17 | OpenAI | 57.7 |
| 18 | xAI | 56.7 |
| 19 | OpenAI | 55.7 |
| 20 | InclusionAI | 54.3 |
| 21 | OpenAI | 48.1 |
Evidence key: Observed
Rows are ordered by the value HealthBench Professional: Evaluating Large Language Models on Real Clinician Chats published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.