Cybergym

Cybergym is a benchmark for how well a model acts as a general-purpose agent. The official description says: "Cybergym - agentic cybersecurity benchmark evaluating autonomous hacking and security task capabilities". Below we describe in simple words what benchmarks of this type generally examine.

Cybergym measures how well a model acts as a general-purpose agent. In detail, it checks:

• solving long, multi-step tasks with many intermediate decisions
• using tools correctly (browsers, terminals, APIs, calendars, other apps)
• adapting when something unexpected happens mid-task
• staying on goal over long task horizons instead of wandering off

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.

Official methodology note: Cybergym - agentic cybersecurity benchmark evaluating autonomous hacking and security task capabilities.

What is the Cybergym benchmark?

Cybergym is a benchmark for how well a model acts as a general-purpose agent. The official description says: 'Cybergym - agentic cybersecurity benchmark evaluating autonomous hacking and security task capabilities'. Below we describe in simple words what benchmarks of this type generally examine.

What does Cybergym test?

Cybergym mainly measures how well a model acts as a general-purpose agent. Concretely, it checks: solving long, multi-step tasks with many intermediate decisions; using tools correctly (browsers, terminals, APIs, calendars, other apps); adapting when something unexpected happens mid-task.

How is the Cybergym score calculated?

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better. The declared maximum score of this benchmark is 100.0.

Where do I get Cybergym data and scores?

The official benchmark page/reference is linked on this page. Every score on Open LLM Radar is also available as open data: download the scores JSON (http://open-llm-radar-api:8000/api/v1/benchmarks/cybergym/scores) or pull it directly from the Open LLM Radar API with any HTTP client.