HealthAgentBench - EHR Data Quality Auditing: leaderboard
HealthAgentBench (2026) evaluates coding agents on healthcare tasks in terminal environments over real clinical data, at xhigh reasoning effort with web browsing disabled. This board is the task success rate in percent over three attempts at auditing an error-injected EHR database, a trial succeeding when every injected error cluster is flagged; higher is better.
Metric: EHR data-quality auditing of an error-injected database (8 tasks; success when every injected error cluster is flagged): task success rate (%) over three attempts per task, a trial succeeding only when it meets the category success criterion; terminal environments in the Harbor framework, xhigh reasoning effort, web browsing disabled. Source: arxiv.org. Saturation forecast: Around May 2027. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 (xHigh) | 33 | #79 (Claude Opus 4.7) |
| 2 | Claude Opus 4.8 (xHigh) | 33 | #56 (Claude Opus 4.8) |
| 3 | GPT-5.5 (xHigh) | 25 | #49 (GPT-5.5) |
| 4 | GPT-5.4 (xHigh) | 12 | #123 (GPT-5.4) |
| 5 | GPT-5.3 Codex (xHigh) | 12 | #116 (GPT-5.3 Codex) |
| 6 | GPT-5.4 Mini (xHigh) | 4 | #389 (GPT-5.4 Mini) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=healthagentbench-ehr-data-quality-auditing · How It Works · Data refreshed daily, snapshot 2026-11-13.