# RemoteExploitBench leaderboard

external benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/external-remote-exploit-bench

RemoteExploitBench (N-day remote exploitation). Every model the catalog carries a published RemoteExploitBench value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | external |
| Measure | N-day target success rate |
| Tasks | Known (N-day) vulnerabilities in deployed applications such as GitLab, PostgreSQL and Jenkins |
| Difficulty | Remote exploit development |
| Published by | [GPT-6 Cyber: Advancing the frontier of aligned cybersecurity agents](https://openai.com/index/introducing-gpt-6-cyber/) |

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.

## RemoteExploitBench ranking

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 | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [GPT-6 Cyber](https://modelchoice.dev/models/gpt-6-cyber) | OpenAI | 94.7 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
