# GPT-6 Cyber

> OpenAI's cyber-specialized version of GPT-6 Astra, released November 5, 2026 to approved Daybreak Red customers for vulnerability discovery, exploit development and web pentesting.

- Maker: OpenAI
- Line: GPT Cyber
- Canonical page: https://ai-tldr.dev/models/gpt-6-cyber/
- Updated: 2026-11-05

## In short

GPT-6 Cyber is OpenAI's cybersecurity model of the GPT-6 generation, released on November 5, 2026. It has a 1.05M-token context window and costs $20.00 per million input tokens and $100.00 per million output tokens. You can use it through OpenAI API (Daybreak Red approval required).

## Facts

| Fact | Value | Source |
| --- | --- | --- |
| Context window | 1.05M (max output 128K) | [1] |
| Input price | $20.00 / 1M tokens | [2] |
| Output price | $100.00 / 1M tokens | [2] |
| Licence | Proprietary · API only | [1] |
| Released | 5 November 2026 | [1] |
| Parameters | Undisclosed | [1] |

## Sources

1. [GPT-6 Cyber System Card (OpenAI Deployment Safety Hub)](https://deploymentsafety.openai.com/gpt-6-cyber)
2. [GPT-6-Cyber model reference — context, pricing, endpoints (OpenAI API docs)](https://developers.openai.com/api/docs/models/gpt-6-cyber)

## Overview

GPT-6 Cyber is OpenAI's cybersecurity model of the GPT-6 generation, released on November 5, 2026. It is a cyber-specialized variant of GPT-6 Astra for authorized security work — vulnerability discovery, exploit development and web pentesting — and OpenAI's system card describes substantially stronger cyber capability and better alignment than GPT-5.6-Cyber. On OpenAI's evaluations it completes 96% of exploit-development requests and 97% of red-teaming requests without refusing.

Access runs through Daybreak Red, OpenAI's tier for advanced, authorized security work. The model starts with a limited group of approved customers through the API and Codex, and OpenAI says it will widen to all Daybreak Red customers over the coming weeks. Organizations must use single sign-on, multi-factor authentication and API IP allowlisting; individuals go through identity and liveness checks, a professional suitability review and a legal attestation, and must use at least one physical FIDO2 security key.

Under OpenAI's Preparedness Framework, GPT-6 Cyber is rated Critical for cybersecurity (the same as GPT-6 Astra), High but not Critical for biological and chemical risk, and below High for AI self-improvement. Extra safeguards target high-risk help with industrial controls, essential services and election systems, developed with Dragos, and a conservative biological safety mode is on for every user. The API lists a 1,050,000-token context window, up to 128,000 output tokens, text and image input, and a price of $20 per million input tokens and $100 per million output tokens.

## Benchmarks

- ExploitGym (intended-exploit success): 48.45% ([source](https://deploymentsafety.openai.com/gpt-6-cyber))
- ExploitGym Hardened: 36.94% ([source](https://deploymentsafety.openai.com/gpt-6-cyber))
- SEC-Bench Pro (pass@1): 87.02% ([source](https://deploymentsafety.openai.com/gpt-6-cyber))
- SRE-Bench (pass@1): 92.84% ([source](https://deploymentsafety.openai.com/gpt-6-cyber))

## Pricing

- Input: $20.00 / 1M tokens
- Output: $100.00 / 1M tokens

Requires Daybreak Red access and organization approval.

Source: https://developers.openai.com/api/docs/models/gpt-6-cyber

## Strengths

- Beats GPT-6 Astra on OpenAI's exploit benchmarks — 48.45% vs 42.43% on ExploitGym and 36.94% vs 29.69% on ExploitGym Hardened
- Higher pass@1 than Astra on SEC-Bench Pro (87.02% vs 85.38%) and SRE-Bench (92.84% vs 87.98%)
- Stays in scope: on the ExploitGym honeypot it went after the out-of-scope target in 0.07% of runs without safety instructions, against 5.51% for GPT-5.6-Cyber
- Irregular's external tests found it solved all 11 CyScenarioBench challenges at least once, against 8 for GPT-6 Astra
- 1.05M-token context window with 128K max output for long agentic security runs

## Best for

- Reach for it when an approved Daybreak Red team needs to investigate a vulnerability, reproduce it in an authorized test environment, or verify a fix.
- Reach for it for long exploit-development and pentest runs in Codex, with sandboxing and Auto-Review, which OpenAI recommends for this model.
- Use Daybreak Blue instead for most defensive work; OpenAI points most defenders there and is widening an alpha of GPT-6 Astra in that tier.
- Keep an eye on its blocker reports: 20.29% of its blocker-reporting responses were flagged in OpenAI's tests, against 11.66% for GPT-6 Astra.

## How to access

- [OpenAI API (Daybreak Red approval required)](https://developers.openai.com/api/docs/models/gpt-6-cyber) — model id `gpt-6-cyber`

## GPT Cyber versions

- GPT-6 Cyber (5 November 2026): Cyber-specialized GPT-6 Astra for Daybreak Red, with a 1.05M-token context window and a Critical cybersecurity rating. — this page
- [GPT-5.6-Cyber](https://ai-tldr.dev/models/gpt-5-6-cyber/) (10 August 2026): Built on GPT-5.6 Sol and trained for zero-day discovery and exploit-chain development; ships in the new Daybreak Red tier with a 400K context window.
- [GPT-5.5-Cyber](https://ai-tldr.dev/models/gpt-5-5-cyber/) (7 May 2026): First GPT Cyber model — more permissive GPT-5.5 tier for vetted defenders under Trusted Access for Cyber; full release June 22, 2026.

## FAQ

### How much does GPT-6 Cyber cost?

OpenAI's API docs list GPT-6 Cyber at $20 per million input tokens and $100 per million output tokens, up from $12.50 and $75 for GPT-5.6-Cyber. Using it needs Daybreak Red access and organization approval; it is not on the open price list for every API account.

### How does GPT-6 Cyber compare with GPT-6 Astra?

In OpenAI's system card GPT-6 Cyber leads GPT-6 Astra on ExploitGym (48.45% vs 42.43%), ExploitGym Hardened (36.94% vs 29.69%), SEC-Bench Pro and SRE-Bench, and sits just behind on ExpertSWE (82.95% vs 83.24%). In Irregular's FrontierCyber test Astra solved slightly more challenges, 96 of 288 against 90.

### Who can get GPT-6 Cyber?

GPT-6 Cyber is limited to approved Daybreak Red customers, starting with a small group and widening to all Daybreak Red customers over the weeks after November 5, 2026. Organizations need SSO, MFA and API IP allowlisting, and individuals must pass identity checks and use a physical FIDO2 security key.

## Links

- [GPT-6 Cyber System Card (OpenAI Deployment Safety Hub)](https://deploymentsafety.openai.com/gpt-6-cyber)
- [GPT-6-Cyber model reference — context, pricing, endpoints (OpenAI API docs)](https://developers.openai.com/api/docs/models/gpt-6-cyber)
- [OpenAI's GPT-6 Cyber ups the ante on vulnerability discovery (CSO Online)](https://www.csoonline.com/article/4244911/openais-gpt-6-cyber-ups-the-ante-on-vulnerability-discovery.html)
- [OpenAI debuts GPT-6 Cyber (CyberScoop)](https://cyberscoop.com/openai-gpt-6-cyber-security-model/)

Source: AI/TLDR LLM registry, https://ai-tldr.dev/models/gpt-6-cyber/