← all models
Flagship Pro & Maxx offers

HacxGPT
Ninja Red

Our most capable offensive-security model.

System overview

Ninja Red is the flagship of the Ninja series: the strongest reasoning stack we operate, tuned end-to-end for offensive security workflows. Where the Lightning line answers quickly, Ninja Red plans, chains tools, and carries multi-stage engagements through to a working artifact.

It is reserved for verified professional tiers because capability cuts both ways. Access is gated by offer level, every request is metered, and acceptable use is enforced contractually — not by a nagging system prompt.

Unrestricted capabilities

Active red-team operationsVulnerability researchAdvanced exploitationMulti-stage agent planningPayload developmentFull-chain engagement support

Make the call

ninja-red · request.sh
curl https://api.hacxgpt.com/v1/chat/completions \
  -H "Authorization: Bearer $HACX_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "ninja-red",
    "reasoning_effort": "medium",
    "messages": [{"role": "user", "content": "..."}]
  }'

Performance metrics

Published scores against frontier peers. Highlighted bars are ours.

Terminal-Bench Hard%
Base v1.019%
▸ HacxGPT Ninja Red40%
TeleCom-Bench%
Base v1.048%
▸ HacxGPT Ninja Red94%
SciCode%
Base v1.042%
▸ HacxGPT Ninja Red42%
Average score%
Base v1.036%
▸ HacxGPT Ninja Red60%

Get started

Run Ninja Red in minutes.

One API key reaches every model on the platform.