Skip to content

Dataset · ArtificialAnalysis

Kubernetes failure scenarios for AI agent diagnostics

ArtificialAnalysis/ITBench-AA

Under 1,000 Kubernetes cluster incident scenarios for evaluating SRE agents.

What's inside

This English text dataset contains public scenarios from the IBM ITBench benchmark (SRE configuration) used on the Artificial Analysis leaderboard. Each record describes a Kubernetes cluster incident with an offline snapshot of alerts, events, traces, and topology, and identifies the faulty component (Deployment, Pod, ConfigMap, etc.).

What you can do with it

  • Evaluating AI agents on automated root cause analysis
  • Testing LLM assistants for DevOps and SRE engineers
Tasks
Question answering
Languages
English
Size
up to 1K rows
Downloads this month
7.7K
Likes
48

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("ArtificialAnalysis/ITBench-AA")