What is the CyberGym leaderboard?

The CyberGym leaderboard ranks 7 AI models based on their performance on this benchmark. Currently, Claude Mythos Preview by Anthropic leads with a score of 0.831. The average score across all models is 0.715.

What is the highest CyberGym score?

The highest CyberGym score is 0.831, achieved by Claude Mythos Preview from Anthropic.

How many models are evaluated on CyberGym?

7 models have been evaluated on the CyberGym benchmark, with 0 verified results and 7 self-reported results.

What categories does CyberGym cover?

CyberGym is categorized under safety, agents, and code. The benchmark evaluates text models.

What is the best open-source model on CyberGym?

GLM-5.1 by Zhipu AI is the top-ranked open-source model on CyberGym, with a score of 0.687 (rank #6).

Which model offers the best value on CyberGym?

Among models scoring within 10% of the leader, GPT-5.5 from OpenAI is the cheapest, at $5.00 per million input tokens with a score of 0.818.

How recent are the CyberGym leaderboard results?

The CyberGym leaderboard was last updated in June 2026 and currently includes 7 evaluated models.

All benchmarks

CyberGym

CyberGym is a benchmark for evaluating AI agents on cybersecurity tasks, testing their ability to identify vulnerabilities, perform security analysis, and complete security-related challenges in a controlled environment.

Claude Mythos Preview from Anthropic currently leads the CyberGym leaderboard with a score of 0.831 across 7 evaluated AI models.