Four-Factor Integrity Breakdown & Primary-Source Evidence
Privacy
Weight 40%
5.5/10
Access Model: Public Prompts / Private Answers
"Access requires manual approval. Do not use references for model training, expose them to an agent during evaluation, or redistribute them."
"The hidden reference is staged only after the agent finishes, so the answer cannot leak into the run."
Originality
Weight 25%
10/10
Reuse Profile: All Original (0% reuse)
"Built with 250+ industry experts and organized by the O*NET / SOC 2018 occupational taxonomy, it spans 55 subdomains across 13 industry clusters and 1,000+ tasks, with 122 active tasks in v1.1."
Freshness
Weight 20%
9/10
Initial Release: 2026-06Latest Refresh: 2026-10Rolling Updates: No
"ALE v1.1 selects 122 active tasks after 30 retirements. Use matching task code, environment profiles, images and data together."
Grading
Weight 15%
3.3/10
Integrity: Self-Reported / Live NetworkSandbox: offline isolated
"This repository is the open evaluation framework: the ale_run toolkit that provisions sandboxes, runs agents, and grades them, plus task implementations across all 55 subdomains"
Benchmark maintainers can embed this live SVG badge in GitHub READMEs or documentation:
[](https://benchaudit.org/benchmarks/agents-last-exam-v1-1/)