Agent's Last Exampass@1

A hard set of desktop and operating-system tasks an AI agent has to finish by looking at the screen and working the machine itself. The score is the share it passes outright — partial credit does not count. Higher is better.

GPT-6.1 Terra by OpenAI holds the best Agent's Last Exam (pass@1) result among tracked models, at 60% (released Oct 26 2026). Higher scores are better on this benchmark.

1. GPT-6.1 Terra (OpenAI) — 60%; 2. GPT-6 Astra (OpenAI) — 59.3%; 3. GPT-6 Sol (OpenAI) — 56.4%; 4. GPT-5.6 Sol (OpenAI) — 52.7%; 5. GPT-5.6 Terra (OpenAI) — 50.4%.

15 tracked models have a published Agent's Last Exam (pass@1) score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.

DeepSeek-V4.1-Flash-1017 by DeepSeek is the highest-ranked model with downloadable weights on Agent's Last Exam (pass@1), scoring 35.4% at rank 8 overall.