# Advanced Cyber Completion Rate — Daybreak Red (ACCR Daybreak Red)

> Share of approved advanced-cyber requests completed rather than refused by GPT-5.6 Cyber under Daybreak Red access.

Canonical page: https://benchlm.ai/benchmarks/advancedcybercompletionratered

- Category: [Instruction Following](/instruction-following)
- Last updated: September 27, 2026

## About ACCR Daybreak Red

- Year: 2026
- Tasks: Internal advanced-cyber request set
- Format: Completion rate
- Difficulty: Cyber access and safeguards
- Paper: [Expanding Daybreak as the cyber defense window narrows](https://openai.com/index/expanding-daybreak-as-the-cyber-defense-window-narrows/)

Daybreak Red provides trusted access to the separate GPT-5.6 Cyber model. ACCR measures access friction and refusals on approved work, not general cyber capability, and remains display only.

ACCR Daybreak Red is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (1 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [GPT-5.6 Cyber](/models/gpt-5-6-cyber) | OpenAI | 95.0% |

## FAQ

### What does ACCR Daybreak Red measure?

Share of approved advanced-cyber requests completed rather than refused by GPT-5.6 Cyber under Daybreak Red access.

### Which model scores highest on ACCR Daybreak Red?

GPT-5.6 Cyber by OpenAI currently leads with a score of 95.0% on ACCR Daybreak Red.

### How many models are evaluated on ACCR Daybreak Red?

1 AI models have been evaluated on ACCR Daybreak Red on BenchLM.

### Does ACCR Daybreak Red affect BenchLM's overall score?

Not directly. ACCR Daybreak Red is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.