# Medical Question Answering on HealthBench Professional Source: [Medical Question Answering on HealthBench Professional]() Collection index: [llms.txt]() Data updated: 2026-05-26T09:08:36.912Z Evaluation of model performance on the Professional subset of the HealthBench dataset for Medical Question Answering, including score, average response length, and comparison delta against MDIA v1.0.40. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Medical Question Answering | | Dataset (including any stated split) | HealthBench Professional | | Metric | Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 7 of 7 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Score](): Higher is better; unit: PERCENT. - [Average Response Length](): Higher is better; unit: SCALAR. - [Delta vs MDIA v1.0.40](): Higher is better; unit: SCALAR. ## Recorded results | Method | Score | Result date | Report paper | | --- | --- | --- | --- | | MDIA v1.0.53 | 62.7 | Not recorded | 6a152939dbd7150506f7e098 | | MDIA v1.0.50 | 61.7 | Not recorded | 6a152939dbd7150506f7e098 | | ChatGPT for Clinicians | 59 | Not recorded | 6a152939dbd7150506f7e098 | | MDIA v1.0.40 | 58.5 | Not recorded | 6a152939dbd7150506f7e098 | | MDIA v1.0.36 | 52.2 | Not recorded | 6a152939dbd7150506f7e098 | | GPT-5.4 base | 48.1 | Not recorded | 6a152939dbd7150506f7e098 | | Physician-written baseline | 43.7 | Not recorded | 6a152939dbd7150506f7e098 | ### MDIA v1.0.53 Score: 62.7. Result ID: 6a15624dfbc373592c0cff5a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"System configuration":"length-guided synthesizer + verifier","Grader model":"GPT-5.4 low"}. This record does not assert independent reproduction. ### MDIA v1.0.50 Score: 61.7. Result ID: 6a15624dfbc373592c0cff5b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"System configuration":"Hydra Platform, full graph endpoint","Grader model":"GPT-5.4 low"}. This record does not assert independent reproduction. ### ChatGPT for Clinicians Score: 59. Result ID: 6a15624dfbc373592c0cff5c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"System configuration":"OpenAI’s best","Grader model":"GPT-5.4 low"}. This record does not assert independent reproduction. ### MDIA v1.0.40 Score: 58.5. Result ID: 6a15624dfbc373592c0cff5d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"System configuration":"multi-turn flatten + 3.1 Pro","Grader model":"GPT-5.4 low"}. This record does not assert independent reproduction. ### MDIA v1.0.36 Score: 52.2. Result ID: 6a15624dfbc373592c0cff5e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"System configuration":"last\_user flatten","Grader model":"GPT-5.4 low"}. This record does not assert independent reproduction. ### GPT-5.4 base Score: 48.1. Result ID: 6a15624dfbc373592c0cff5f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"System configuration":"single-agent","Grader model":"GPT-5.4 low"}. This record does not assert independent reproduction. ### Physician-written baseline Score: 43.7. Result ID: 6a15624dfbc373592c0cff60. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Grader model":"GPT-5.4 low"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.