Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
external benchmark
RemoteExploitBench (N-day remote exploitation). Every model the catalog carries a published RemoteExploitBench value for, ranked by that value.
OpenAI's benchmark of whether a model can build a working exploit for a known vulnerability and use it against a remote target, on real applications such as GitLab, PostgreSQL and Jenkins deployed in a controlled environment. One model in the ModelChoice catalog has a published value.
Each point is a model with a published RemoteExploitBench value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
1 model with a published RemoteExploitBench value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | N-day target success rate |
|---|---|---|---|
| 1 | OpenAI | 94.7 |
Evidence key: Observed
Rows are ordered by the value GPT-6 Cyber: Advancing the frontier of aligned cybersecurity agents published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
What the benchmark measures and how its value is published, from the source’s own record.
RemoteExploitBench (RemoteExploitBench (N-day remote exploitation)) is an external benchmark published by GPT-6 Cyber: Advancing the frontier of aligned cybersecurity agents. OpenAI's benchmark of whether a model can build a working exploit for a known vulnerability and use it against a remote target, on real applications such as GitLab, PostgreSQL and Jenkins deployed in a controlled environment. Its evaluation set is described by the source as known (N-day) vulnerabilities in deployed applications such as GitLab, PostgreSQL and Jenkins. The source rates its difficulty as remote exploit development.
Each model's result is published by GPT-6 Cyber: Advancing the frontier of aligned cybersecurity agents as n-day target success rate. ModelChoice prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 1 model that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.