safety benchmark
ExploitBench Leaderboard
ExploitBench is a cybersecurity benchmark that evaluates a model's ability to discover and exploit software vulnerabilities, reported as the fraction of challenges where the model captures the target (Cap%).
Updated Aug 17, 2026
Models5
Model coverage5
MetricScore
EvidenceB
ExploitBench Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAN | Score78.0% | Percentile100.0% | Participants5 | EvidenceC | Evaluated |
| Rank02 | ModelOP | Score73.5% | Percentile75.0% | Participants5 | EvidenceC | Evaluated |
| Rank03 | ModelZA | Score54.4% | Percentile50.0% | Participants5 | EvidenceC | Evaluated |
| Rank04 | ModelOP | Score52.9% | Percentile25.0% | Participants5 | EvidenceC | Evaluated |
| Rank05 | ModelOP | Score33.2% | Percentile0.0% | Participants5 | EvidenceC | Evaluated |
ExploitBench Highlights
The leading models and scores on this benchmark.
ExploitBench Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for ExploitBench
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis exploitbench 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.
- 01ANClaude Fable 5Anthropic
- Score
- 78.0%
- Price
- $10 input / $50 output per 1M tokens
- Speed
- Up to 9.1 tok/s via Anthropic
Strengths
- Ranks #1 of 5 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures ExploitBench, not total model capability
- 02OPGPT-5.6 SolOpenAI
- Score
- 73.5%
- Price
- $5.0 input / $30 output per 1M tokens
- Speed
- Up to 27 tok/s via OpenAI
Strengths
- Ranks #2 of 5 compared models
- 75th percentile on this benchmark
- C evidence result
Considerations
- This result measures ExploitBench, not total model capability
- 03ZAGLM-5.3Zhipu AI
- Score
- 54.4%
Strengths
- Ranks #3 of 5 compared models
- 50th percentile on this benchmark
- C evidence result
Considerations
- This result measures ExploitBench, not total model capability
- 04OPGPT-5.6 TerraOpenAI
- Score
- 52.9%
- Price
- $2.0 input / $12 output per 1M tokens
- Speed
- Up to 35 tok/s via OpenAI
Strengths
- Ranks #4 of 5 compared models
- 25th percentile on this benchmark
- C evidence result
Considerations
- This result measures ExploitBench, not total model capability
- 05OPGPT-5.6 LunaOpenAI
- Score
- 33.2%
- Price
- $0.20 input / $1.2 output per 1M tokens
- Speed
- Up to 49 tok/s via OpenAI
Strengths
- Ranks #5 of 5 compared models
- 0th percentile on this benchmark
- C evidence result
Considerations
- This result measures ExploitBench, not total model capability
Selection summary
Best AI Models for ExploitBench
Claude Fable 5 currently leads ExploitBench with 78.0%. 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 ExploitBench?
What ExploitBench measures and how its scores work.
ExploitBench is a cybersecurity benchmark that evaluates a model's ability to discover and exploit software vulnerabilities, reported as the fraction of challenges where the model captures the target (Cap%).
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- ExploitBench
- Modality
- text
- Primary category
- safety
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- exploitbench|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about ExploitBench.
Which model scores highest on ExploitBench?
Claude Fable 5 is currently ranked first with 78.0%.
What does ExploitBench measure?
ExploitBench is a cybersecurity benchmark that evaluates a model's ability to discover and exploit software vulnerabilities, reported as the fraction of challenges where the model captures the target (Cap%).
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
5 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.