safety benchmark
CyberGym Leaderboard
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.
Updated Aug 17, 2026
Models13
Model coverage13
MetricScore
EvidenceB
CyberGym Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelZA | Score84.5% | Percentile100.0% | Participants13 | EvidenceC | Evaluated |
| Rank02 | ModelDE | Score83.3% | Percentile91.7% | Participants13 | EvidenceC | Evaluated |
| Rank03 | ModelGO | Score83.2% | Percentile83.3% | Participants13 | EvidenceC | Evaluated |
| Rank04 | ModelAN | Score83.1% | Percentile75.0% | Participants13 | EvidenceC | Evaluated |
| Rank05 | ModelOP | Score81.8% | Percentile66.7% | Participants13 | EvidenceC | Evaluated |
| Rank06 | ModelAN | Score78.8% | Percentile58.3% | Participants13 | EvidenceC | Evaluated |
| Rank07 | ModelDE | Score76.7% | Percentile50.0% | Participants13 | EvidenceC | Evaluated |
| Rank08 | ModelAN | Score73.8% | Percentile41.7% | Participants13 | EvidenceC | Evaluated |
| Rank09 | ModelAN | Score73.1% | Percentile33.3% | Participants13 | EvidenceC | Evaluated |
| Rank10 | ModelBY | Score70.2% | Percentile25.0% | Participants13 | EvidenceC | Evaluated |
| Rank11 | ModelZA | Score68.7% | Percentile16.7% | Participants13 | EvidenceC | Evaluated |
| Rank12 | ModelBY | Score67.0% | Percentile8.3% | Participants13 | EvidenceC | Evaluated |
| Rank13 | ModelMA | Score41.3% | Percentile0.0% | Participants13 | EvidenceC | Evaluated |
CyberGym Highlights
The leading models and scores on this benchmark.
CyberGym Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for CyberGym
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis cybergym AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01ZAGLM-5.3Zhipu AI
- Score
- 84.5%
Strengths
- Ranks #1 of 13 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures CyberGym, not total model capability
- 02DEDeepSeek-V4-Pro-0813DeepSeek
- Score
- 83.3%
- Price
- $0.43 input / $0.87 output per 1M tokens
- Speed
- Up to 33 tok/s via DeepSeek
Strengths
- Ranks #2 of 13 compared models
- 92th percentile on this benchmark
- C evidence result
Considerations
- This result measures CyberGym, not total model capability
- 03GOGemini 3.5 Flash CyberGoogle
- Score
- 83.2%
Strengths
- Ranks #3 of 13 compared models
- 83th percentile on this benchmark
- C evidence result
Considerations
- This result measures CyberGym, not total model capability
- 04ANClaude Mythos PreviewAnthropic
- Score
- 83.1%
Strengths
- Ranks #4 of 13 compared models
- 75th percentile on this benchmark
- C evidence result
Considerations
- This result measures CyberGym, not total model capability
- 05OPGPT-5.5OpenAI
- Score
- 81.8%
- Price
- $5.0 input / $30 output per 1M tokens
Strengths
- Ranks #5 of 13 compared models
- 67th percentile on this benchmark
- C evidence result
Considerations
- This result measures CyberGym, not total model capability
Selection summary
Best AI Models for CyberGym
GLM-5.3 currently leads CyberGym with 84.5%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is CyberGym?
What CyberGym measures and how its scores work.
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.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- CyberGym
- Modality
- text
- Primary category
- safety
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- cybergym|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about CyberGym.
Which model scores highest on CyberGym?
GLM-5.3 is currently ranked first with 84.5%.
What does CyberGym measure?
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.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
13 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.