Instructions to use aashishk029/shinobu-7b-v8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use aashishk029/shinobu-7b-v8 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="aashishk029/shinobu-7b-v8",
	filename="shinobu-7b-v8.Q4_K_M.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use aashishk029/shinobu-7b-v8 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 aashishk029/shinobu-7b-v8:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf aashishk029/shinobu-7b-v8:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf aashishk029/shinobu-7b-v8:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf aashishk029/shinobu-7b-v8: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 aashishk029/shinobu-7b-v8:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf aashishk029/shinobu-7b-v8: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 aashishk029/shinobu-7b-v8:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf aashishk029/shinobu-7b-v8:Q4_K_M

Use Docker

docker model run hf.co/aashishk029/shinobu-7b-v8:Q4_K_M

LM Studio
Jan

vLLM

How to use aashishk029/shinobu-7b-v8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "aashishk029/shinobu-7b-v8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "aashishk029/shinobu-7b-v8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/aashishk029/shinobu-7b-v8:Q4_K_M

Ollama
How to use aashishk029/shinobu-7b-v8 with Ollama:
```
ollama run hf.co/aashishk029/shinobu-7b-v8:Q4_K_M
```

Unsloth Studio

How to use aashishk029/shinobu-7b-v8 with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for aashishk029/shinobu-7b-v8 to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for aashishk029/shinobu-7b-v8 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for aashishk029/shinobu-7b-v8 to start chatting

How to use aashishk029/shinobu-7b-v8 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf aashishk029/shinobu-7b-v8:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "aashishk029/shinobu-7b-v8:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use aashishk029/shinobu-7b-v8 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf aashishk029/shinobu-7b-v8:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default aashishk029/shinobu-7b-v8:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use aashishk029/shinobu-7b-v8 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf aashishk029/shinobu-7b-v8:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "aashishk029/shinobu-7b-v8:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use aashishk029/shinobu-7b-v8 with Docker Model Runner:
```
docker model run hf.co/aashishk029/shinobu-7b-v8:Q4_K_M
```

Lemonade

How to use aashishk029/shinobu-7b-v8 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull aashishk029/shinobu-7b-v8:Q4_K_M

Run and chat with the model

lemonade run user.shinobu-7b-v8-Q4_K_M

List all available models

lemonade list

Shinobu 7B v8 — Q4_K_M GGUF

Production cybersecurity-triage copilot for the Shinobu on-prem security stack.

Base: Qwen2.5-7B-Instruct + Shinobu LoRA v8 (merged + Q4_K_M quantized). Size: 4.4 GB. Runs on 16 GB RAM / 8-core CPU. No GPU required. Quality: 93% verdict accuracy on 30 held-out hard triage probes (FP 8/10, TP 10/10, abstain 10/10).

Quick install via ollama

curl -fL -o shinobu-7b-v8.Q4_K_M.gguf \
  https://huggingface.co/aashishk029/shinobu-7b-v8/resolve/main/shinobu-7b-v8.Q4_K_M.gguf
sha256sum -c <(curl -s https://huggingface.co/aashishk029/shinobu-7b-v8/resolve/main/SHA256SUMS)

cat > Modelfile <<'MF'
FROM ./shinobu-7b-v8.Q4_K_M.gguf
TEMPLATE """{{- if .System }}<|im_start|>system
{{ .System }}<|im_end|>
{{ end }}{{- range .Messages }}<|im_start|>{{ .Role }}
{{ .Content }}<|im_end|>
{{ end }}<|im_start|>assistant
"""
PARAMETER stop "<|im_start|>"
PARAMETER stop "<|im_end|>"
PARAMETER temperature 0
PARAMETER num_ctx 4096
SYSTEM """You are Shinobu, an on-prem cybersecurity assistant. Triage alerts: identify the attack, map to MITRE ATT&CK, judge true/false positive, set severity, and recommend a response action that the operator must APPROVE before execution (never auto-execute)."""
MF

ollama create shinobu-v8 -f Modelfile
ollama run shinobu-v8 "Triage this: 380 failed SSH logins from 91.219.236.140 to db-prod-01..."

Full deployment

Use the Shinobu installer for systemd + audit chain + approval dashboard:

git clone https://github.com/aashishk029/shinobu
cd shinobu
sudo SHINOBU_GGUF_URL='https://huggingface.co/aashishk029/shinobu-7b-v8/resolve/main/shinobu-7b-v8.Q4_K_M.gguf' \
     SHINOBU_GGUF_SHA256='f1ae7e1c2581706cd514e661c30b4bf23aae020ccce9e34d2cad4d449845a5fd' \
     bash install.sh

Then follow FIRSTRUN.md.

Honest limits

AI is L2 advisory, not L1 detection. Every state-changing action requires a human click.
Atomic Red Team fusion recall = 46% on 1,800 events. Improves with pilot-driven rule expansion, not magic.
Abstains rather than fabricating verdicts on thin telemetry. This is intentional and the v8 calibration win.

License

Source-available under Polyform Free Trial 1.0.0 for non-commercial / free trial use. Paid commercial subscription required for production deployment in for-profit contexts. See LICENSE, PRICING.md, and TRIAL_TERMS.md.

Base model Qwen2.5-7B-Instruct is governed by Alibaba's Tongyi Qianwen License (see NOTICE).

Downloads last month: 2

GGUF

Model size

8B params

Architecture

qwen2

Hardware compatibility

4-bit

Model tree for aashishk029/shinobu-7b-v8

Base model

Qwen/Qwen2.5-7B

Finetuned

Qwen/Qwen2.5-7B-Instruct

Adapter

(2372)

this model