Instructions to use nyxspecter4/kinetigor-dpo-cybersec-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use nyxspecter4/kinetigor-dpo-cybersec-gguf with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M

Use Docker

docker model run hf.co/nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M

LM Studio
Jan

vLLM

How to use nyxspecter4/kinetigor-dpo-cybersec-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nyxspecter4/kinetigor-dpo-cybersec-gguf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nyxspecter4/kinetigor-dpo-cybersec-gguf",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M

Ollama
How to use nyxspecter4/kinetigor-dpo-cybersec-gguf with Ollama:
```
ollama run hf.co/nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use nyxspecter4/kinetigor-dpo-cybersec-gguf with Docker Model Runner:
```
docker model run hf.co/nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M
```

Lemonade

How to use nyxspecter4/kinetigor-dpo-cybersec-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M

Run and chat with the model

lemonade run user.kinetigor-dpo-cybersec-gguf-Q4_K_M

List all available models

lemonade list

Atomic Chat

KIN v6 DPO — Cybersecurity AI (GGUF)

GGUF quantizations of KIN v6 DPO, a cybersecurity AI fine-tuned via DPO on Qwen2.5-0.5B-Instruct. Run locally with Ollama, llama.cpp, or any GGUF-compatible runtime.

Available Quantizations

File	Quant	Size	Use Case
`kinetigor-v6-Q4_K_M.gguf`	Q4_K_M	~398 MB	Best balance — recommended for most users
`kinetigor-v6-Q8_0.gguf`	Q8_0	~531 MB	Highest quality, near-lossless

Quick Start — Ollama

# Pull and run directly from Hugging Face
ollama run hf.co/nyxspecter4/kinetigor-dpo-cybersec-gguf:Q4_K_M

# Or pull the Q8_0 version
ollama run hf.co/nyxspecter4/kinetigor-dpo-cybersec-gguf:Q8_0

Quick Start — llama.cpp

# Build llama.cpp
git clone https://github.com/ggml-org/llama.cpp
cd llama.cpp && make

# Run KIN
./llama-cli -m kinetigor-v6-Q4_K_M.gguf \
  -p "How do I detect a foothold after a phishing attack?" \
  --system-prompt "You are KIN — a sharp cybersecurity AI partner. Direct, opinionated, specific. Name tools, CVEs, companies. Sound like a senior engineer at a bar, not a textbook."

Critical: System Prompt

KIN was trained with a specific system prompt. Using a different prompt will degrade quality significantly.

You are KIN — a sharp cybersecurity AI partner. Direct, opinionated, specific. Name tools, CVEs, companies. Sound like a senior engineer at a bar, not a textbook. Lead with your boldest take. End with a specific action. Max 2-3 paragraphs. Open with your take, not your title. No "As a cybersecurity AI expert." Name products: "CrowdStrike Falcon" not "use EDR". "Duo push MFA" not "implement MFA".

About KIN

KIN answers security questions like a senior engineer at a bar — direct, opinionated, and specific. It names real tools (CrowdStrike Falcon, Velociraptor, Duo MFA, KnowBe4), references real CVEs (CVE-2023-4863, CVE-2021-44228, CVE-2024-3094), and knows real incidents (MGM, Colonial Pipeline, NotPetya, Maersk $300M, Merck $670M).

Training

Base model: Qwen2.5-0.5B-Instruct
Fine-tuning: DPO (Direct Preference Optimization)
Base model repo: nyxspecter4/kinetigor-dpo-cybersec
Training data: nyxspecter4/kin-dpo-data

License

Apache 2.0

Downloads last month: 106

GGUF

Model size

0.5B params

Architecture

qwen2

Hardware compatibility

4-bit

8-bit

Model tree for nyxspecter4/kinetigor-dpo-cybersec-gguf

Base model

mistralai/Mistral-7B-Instruct-v0.2

Finetuned

nyxspecter4/kinetigor-dpo-cybersec

Quantized

(1)

this model