← Greps
- HF Model
- vxkyyy/vyber-security-7b-gguf
- Base Model
- Qwen/Qwen2.5-7B-Instruct
- Pipeline
- text-generation
- Downloads
- 232
- Likes
- 1
- cybersecurity
- qwen2.5
- lora
- gguf
- instruct
cyber-llm
vyber-security-7b — Qwen2.5 Cybersecurity LoRA
Uploaded to Hugging Face by vxkyyy — Qwen2.5 7B Instruct LoRA packaged as GGUF for local cybersecurity chat workloads.
What Is This?
Uploaded to Hugging Face by vxkyyy. vyber-security-7b is a LoRA adapter on Qwen/Qwen2.5-7B-Instruct, distributed as GGUF for llama.cpp-compatible inference. A smaller 1.5B variant also exists in the same namespace.
Metadata
| Field | Value |
|---|---|
| Author | vxkyyy |
| Parameter scale | 7B (1.5B sibling available) |
| Format | GGUF |
| Downloads | ~232 |
| License | Apache 2.0 |
Why It Might Matter
Lightweight cyber-LLMs fill a practical gap — runnable on laptops and edge devices for:
- Offline security Q&A during engagements
- Air-gapped lab environments without API dependencies
- Rapid adapter experimentation (merge LoRA, swap bases)
The 7B scale is the sweet spot for comparing domain fine-tune lift without MoE infrastructure overhead.
Evaluation Suggestions
- MITRE technique explanation accuracy
- False positive rate on benign vs. malicious code snippets
- Refusal behavior on clearly unauthorized attack instructions
Attribution
Open-source model indexed for security research inquiry. Uploaded to Hugging Face by vxkyyy.