Downloads · 30 days
4.8K
50% of all-time downloads
Aapolon/Qwen3.5-9B-EmperoAI-Qwen3.8-Distill-Heretic-CyberStrike-GGUF
Qwen3.5-9B-EmperoAI-Qwen3.8-Distill-Heretic-CyberStrike-GGUF is a machine learning model from Aapolon. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for gguf. The card lists the license as apache-2.0.
⚠️ IMPORTANT: This model is explicitly designed for authorized security research, offensive cybersecurity, and red teaming. By using this model, you agree to strictly adhere to the Responsible Use Guidelines outlined…
Downloads · 30 days
4.8K
50% of all-time downloads
All-time downloads
9.6K
Public
Repo size
13.3 GB
Likes
4
Public
Click a slice to open those files.
.gguf13.3 GB · 100%
From the Hugging Face model README
⚠️ IMPORTANT: This model is explicitly designed for authorized security research, offensive cybersecurity, and red teaming. By using this model, you agree to strictly adhere to the Responsible Use Guidelines outlined below.
Qwen3.5-9B-Qwen3.8-Distill-Heretic-CyberStrike-GGUF is a highly specialized, fine-tuned large language model optimized for offensive cybersecurity, vulnerability research, and authorized penetration testing.
Built upon the petruhonk/Qwen3.8-9B-Distill-uncensored-heretic base model and fine-tuned on the extensive CyberStrike SFT 120K dataset, this GGUF-quantized model delivers expert-level, actionable technical depth without artificial safety refusals. It is explicitly designed to serve as the cognitive engine for the CyberStrike AI Framework.
This model is purpose-built for security professionals and researchers. Recommended use cases include:
| Component | Details |
|---|---|
| Base Model | petruhonk/Qwen3.8-9B-Distill-uncensored-heretic |
| Fine-Tuning Dataset | oyildirim/cyberstrike-sft-120k (121,422 expert-level red team instruction-response pairs) |
| Format | GGUF (Optimized for efficient CPU/GPU inference via llama.cpp, Ollama, or LM Studio) |
| Domains Covered | Cloud Security, CVE Reasoning, CIS Hardening, MITRE ATT&CK, OWASP (Web/ASVS), NIST, and General Offensive Operations |
| Behavior | Uncensored, operator-level technical detail, framework-aligned, no generic refusals |