Agentic computer use

Agent's Last Examscore

The graded score on the same desktop and operating-system tasks in Agent's Last Exam, giving partial credit for progress beyond the strict pass-or-fail result. Higher is better.

Rankings

Higher is better

Agent's Last Exam — frequently asked questions

What is Agent's Last Exam?
The graded score on the same desktop and operating-system tasks in Agent's Last Exam, giving partial credit for progress beyond the strict pass-or-fail result. Higher is better.
Which AI model scores highest on Agent's Last Exam?
Qwen3.8-27B by Qwen holds the best Agent's Last Exam (score) result among tracked models, at 42.9% (released Aug 14 2026). Higher scores are better on this benchmark.
What are the top 1 models on Agent's Last Exam?
1. Qwen3.8-27B (Qwen) — 42.9%.
How many models have a published Agent's Last Exam score?
1 tracked model has a published Agent's Last Exam (score) score. Scores are the figures reported by each lab at that model's release, so this page is a record of results over time rather than a re-run leaderboard.
What is the best open model on Agent's Last Exam?
Qwen3.8-27B by Qwen is the highest-ranked model with downloadable weights on Agent's Last Exam (score), scoring 42.9% at rank 1 overall.
← All benchmarks