safety benchmark
CyberSecEval 4
CyberSecEval 4 is an evaluation suite covering cybersecurity-related capabilities and risks of large language models. The insecure-code-generation tracks measure whether a model produces vulnerable code: the Instruct track presents coding requests designed to elicit known insecure patterns, while the Autocomplete track prompts the model with code context leading up to a known insecure pattern, with vulnerabilities detected via static analysis.
Updated Aug 11, 2026
Models1
Model coverage1
MetricScore
EvidenceB
CyberSecEval 4 Ranking
Higher score ranks better on this benchmark.
| 01 | MI | 63.0% | 100.0% | 1 | C |
CyberSecEval 4 Highlights
The leading models and scores on this benchmark.
What is CyberSecEval 4?
What CyberSecEval 4 measures and how its scores work.
CyberSecEval 4 is an evaluation suite covering cybersecurity-related capabilities and risks of large language models. The insecure-code-generation tracks measure whether a model produces vulnerable code: the Instruct track presents coding requests designed to elicit known insecure patterns, while the Autocomplete track prompts the model with code context leading up to a known insecure pattern, with vulnerabilities detected via static analysis.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- CyberSecEval 4
- Modality
- text
- Primary category
- safety
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- cyberseceval-4|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about CyberSecEval 4.
Which model scores highest on CyberSecEval 4?
MAI-Thinking-1 is currently ranked first with 63.0%.
What does CyberSecEval 4 measure?
CyberSecEval 4 is an evaluation suite covering cybersecurity-related capabilities and risks of large language models. The insecure-code-generation tracks measure whether a model produces vulnerable code: the Instruct track presents coding requests designed to elicit known insecure patterns, while the Autocomplete track prompts the model with code context leading up to a known insecure pattern, with vulnerabilities detected via static analysis.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 model results are currently shown.
Does this benchmark affect the overall score?
No. This benchmark is shown for reference but does not contribute to the overall score.