# Agents' Last Exam v1.1 Contamination Audit

> Agents' Last Exam v1.1 scores 7/10 (Tier A, Low Risk) on BenchAudit as of November 2026, reflecting public prompts with private held-out answers and self-reported or live-network evaluation.

- **Canonical URL:** https://benchaudit.org/benchmarks/agents-last-exam-v1-1/
- **Benchmark ID:** `agents-last-exam-v1-1`
- **Hub:** UC Berkeley RDI
- **Category:** Knowledge Work
- **Domains:** Business, Finance, Legal, Medical, Science, Coding
- **Skills:** Agentic, Tool Use, Reasoning, Knowledge Work, Coding, Multimodal, Instruction Following
- **BenchAudit Score:** 7/10 (Tier A - Low Risk)
- **Last Audited:** 2026-11-19 (Ref 2026-11)

## Factor Scores

| Factor | Score (/10) | Weight |
| --- | ---: | ---: |
| Privacy | 5.5 | 40% |
| Originality | 10 | 25% |
| Freshness | 9 | 20% |
| Grading | 3.3 | 15% |

## Primary-Source Evidence

### Privacy
> "Access requires manual approval. Do not use references for model training, expose them to an agent during evaluation, or redistribute them."
- Source: https://huggingface.co/datasets/agents-last-exam/agents-last-exam-reference

### Originality
> "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."
- Source: https://agents-last-exam.org/docs/ale/index.html

### Freshness
> "ALE v1.1 selects 122 active tasks after 30 retirements. Use matching task code, environment profiles, images and data together."
- Source: https://raw.githubusercontent.com/rdi-berkeley/agents-last-exam/main/docs/releases/v1.1.md

### Grading
> "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"
- Source: https://github.com/rdi-berkeley/agents-last-exam
