Instructions to use Tarun58/VaultCoder-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Tarun58/VaultCoder-3B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Tarun58/VaultCoder-3B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Tarun58/VaultCoder-3B", dtype="auto")

llama-cpp-python

How to use Tarun58/VaultCoder-3B with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Tarun58/VaultCoder-3B",
	filename="VaultCoder-3B-SDLC.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 Tarun58/VaultCoder-3B with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Tarun58/VaultCoder-3B
# Run inference directly in the terminal:
llama-cli -hf Tarun58/VaultCoder-3B

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Tarun58/VaultCoder-3B
# Run inference directly in the terminal:
llama-cli -hf Tarun58/VaultCoder-3B

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 Tarun58/VaultCoder-3B
# Run inference directly in the terminal:
./llama-cli -hf Tarun58/VaultCoder-3B

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 Tarun58/VaultCoder-3B
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Tarun58/VaultCoder-3B

Use Docker

docker model run hf.co/Tarun58/VaultCoder-3B

LM Studio
Jan

vLLM

How to use Tarun58/VaultCoder-3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Tarun58/VaultCoder-3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Tarun58/VaultCoder-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Tarun58/VaultCoder-3B

SGLang

How to use Tarun58/VaultCoder-3B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Tarun58/VaultCoder-3B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Tarun58/VaultCoder-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Tarun58/VaultCoder-3B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Tarun58/VaultCoder-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use Tarun58/VaultCoder-3B with Ollama:
```
ollama run hf.co/Tarun58/VaultCoder-3B
```

Unsloth Studio

How to use Tarun58/VaultCoder-3B 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 Tarun58/VaultCoder-3B 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 Tarun58/VaultCoder-3B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Tarun58/VaultCoder-3B to start chatting

How to use Tarun58/VaultCoder-3B with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Tarun58/VaultCoder-3B

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": "Tarun58/VaultCoder-3B"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Tarun58/VaultCoder-3B with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Tarun58/VaultCoder-3B

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 Tarun58/VaultCoder-3B

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use Tarun58/VaultCoder-3B with Docker Model Runner:
```
docker model run hf.co/Tarun58/VaultCoder-3B
```

Lemonade

How to use Tarun58/VaultCoder-3B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Tarun58/VaultCoder-3B

Run and chat with the model

lemonade run user.VaultCoder-3B-{{QUANT_TAG}}

List all available models

lemonade list

🛡️ VaultCoder-3B-SDLC

VaultCoder-3B-SDLC is a hyper-specialized, privacy-first Large Language Model (LLM) engineered for high-stakes enterprise architecture and secure software development lifecycle (SDLC) management.

While generic models struggle with the nuances of regulated industries, VaultCoder has been fine-tuned using professional-grade Knowledge Distillation to provide elite-level performance in local environments.

🏛️ Industry-Specific Specializations

VaultCoder has been meticulously trained across 7 high-value sectors:

🏥 Healthcare: HIPAA-compliant AWS/Azure architectures, HL7 FHIR data mapping, and secure patient portal logic.
🏘️ Real Estate: RESO Web API integration, automated MLS sync services, and complex financial/mortgage calculators.
🏎️ Automobile: Real-time OBD-II telematics processing, EV fleet management APIs, and sensor data integration.
🎨 AR/3D Design: WebXR augmented reality components, 360° image-to-2D floor plan generation, and Unity 3D scripts.
🎓 EdTech: LMS database schemas (SCORM/xAPI) and AI-driven pedagogy for student feedback.
📈 Sales Ops: Predictive lead scoring engines and NLP-based automated CRM outreach.
⚖️ Strategic Advisory: Automated M&A due diligence (RAG), Monte Carlo risk simulations, and SWOT analysis.

🛠️ Training Methodology

VaultCoder-3B-SDLC was developed using a state-of-the-art Knowledge Distillation pipeline:

Teacher Model: Qwen2.5-Coder-7B-Instruct (High-reasoning "Pro" architect).
Dataset: A proprietary 24-record "Gold Standard" dataset generated with deep architectural reasoning and zero placeholders.
Technique: LoRA (Low-Rank Adaptation) fine-tuning on the Qwen2.5-Coder-3B base.
Quantization: 4-bit GGUF (Q4_K_M) for maximum speed and minimum RAM usage on standard CPUs.

🚀 Quick Start (Local Deployment)

VaultCoder is optimized for private, local execution using Ollama.

1. Create a Modelfile

Create a file named Modelfile:

FROM ./VaultCoder-3B-SDLC.gguf

SYSTEM """
You are VaultCoder, a World-Class Software Engineer and Strategic Business Consultant. 
Your goal is to provide production-ready, secure, and industry-optimized solutions.
"""

PARAMETER temperature 0.3
PARAMETER num_ctx 8192

2. Run the Model

ollama create vaultcoder -f Modelfile
ollama run vaultcoder

🔒 Privacy & Compliance

VaultCoder is designed to run 100% offline. Your intellectual property, patient data, and financial records never leave your local infrastructure, making it the ideal choice for companies that cannot use cloud-based AI providers.