ADeptS-Bench - Task Success (Desktop): leaderboard
ADeptS-Bench (2026) tests computer-use agents on single decision points in mobile and desktop screenshots. This board is the share of benign desktop safety-stream tasks the agent completes with the correct action, in percent; higher is better.
Metric: Task success rate (%; share of the 501 benign desktop tasks completed with the correct next action; paired benign and malicious GUI tasks with visual threats (phishing overlays, misleading buttons, prompt injection), graded on the next action against ground-truth boxes; mean of 10 runs at temperature 0.1). Source: arxiv.org. Saturation forecast: Around February 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 235B A22B | 86.6 | #479 |
| 2 | Gemini 3.1 Pro (Preview) | 82.9 | #94 |
| 3 | Claude Opus 4.7 | 78.1 | #79 |
| 4 | GPT-5.4 | 68.8 | #123 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=adepts-bench-task-success-desktop · How It Works · Data refreshed daily, snapshot 2026-11-13.