Agents' Last Exam
A challenging public evaluation benchmark targeting the comprehensive capabilities of AI agents, assessing models on planning, execution, and correction in long-horizon, multi-step autonomous task scenarios.
Timeline (last 90 days)
Tev-27B, trained only on ALE-related data, can nonetheless generalize to Terminal-Bench, suggesting that it has learned some domain-agnostic long-horizon task execution capability rather than merely memorizing specific task solutions.
Tev-27B was trained entirely with PPRS on the computational and mathematical sciences domains of ALE (Agents' Last Exam), matching GPT-6 Astra and Claude Opus 5 on the Pass@5 metric of ALE-CM.
GLM-5.3 achieved a state-of-the-art score among open-source models on the Agents' Last Exam benchmark.
All Facts (3)
Tev-27B, trained only on ALE-related data, can nonetheless generalize to Terminal-Bench, suggesting that it has learned some domain-agnostic long-horizon task execution capability rather than merely memorizing specific task solutions.
50%UnverifiedTev-27B was trained entirely with PPRS on the computational and mathematical sciences domains of ALE (Agents' Last Exam), matching GPT-6 Astra and Claude Opus 5 on the Pass@5 metric of ALE-CM.
50%UnverifiedGLM-5.3 achieved a state-of-the-art score among open-source models on the Agents' Last Exam benchmark.
50%