DTap-Bench (Benign Tasks): leaderboard

Metric: Benign task success rate (%; 2,806 benign agent tasks, no attack applied; a task succeeds when a rule-based check of the final environment state finds the instruction fully completed; mean of the 14 domain rates). Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.

Top models

#ModelScore
1Gemini 3 Pro87
2GPT-5.485.3
3GPT-5.280.5
4GPT-5.178.7
5GPT-OSS-120B36.2

Interactive version: theaggregate.ai/benchmark?slug=dtap-bench-benign-tasks · How It Works · Data refreshed daily, snapshot 2026-09-26.