AJ-Bench (Agent-as-a-Judge) - Word: leaderboard
Metric: F1 (%) on Word GUI tasks (OSWorld; 12 tasks, 24 trajectories), with the judge running as an MCPMark agent that calls tools and inspects the live environment for evidence, of AJ-Bench, judging whether agent trajectories succeeded against binary human-verified labels (155 tasks, 516 trajectories), default model configurations; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V3.2 | 78.64 |
| 2 | DeepSeek V3.2 (Thinking) | 78.57 |
| 3 | GPT-5 Mini (Low) | 72.22 |
| 4 | GPT-5 Mini | 72 |
| 5 | GPT-5 Mini (High) | 63.64 |
Interactive version: theaggregate.ai/benchmark?slug=aj-bench-agent-as-a-judge-word · How It Works · Data refreshed daily, snapshot 2026-10-07.