DTap-Bench (Benign Tasks) - Medical: leaderboard
Metric: Benign task success rate (%; 642 benign medical service tasks, no attack applied; a task succeeds when a rule-based check of the final environment state finds the instruction fully completed; mean of its task-category rates). Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 | 86.1 |
| 2 | Gemini 3 Pro | 81.9 |
| 3 | GPT-5.2 | 66.3 |
| 4 | GPT-5.1 | 52.8 |
| 5 | GPT-OSS-120B | 35.8 |
Interactive version: theaggregate.ai/benchmark?slug=dtap-bench-benign-tasks-medical · How It Works · Data refreshed daily, snapshot 2026-09-26.