Instructions to use LADDOO22212015/Vanguard-8B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LADDOO22212015/Vanguard-8B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="LADDOO22212015/Vanguard-8B-GGUF",
	filename="Vanguard-8B-Merged-BF16.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 LADDOO22212015/Vanguard-8B-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 LADDOO22212015/Vanguard-8B-GGUF:BF16
# Run inference directly in the terminal:
llama cli -hf LADDOO22212015/Vanguard-8B-GGUF:BF16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf LADDOO22212015/Vanguard-8B-GGUF:BF16
# Run inference directly in the terminal:
llama cli -hf LADDOO22212015/Vanguard-8B-GGUF:BF16

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 LADDOO22212015/Vanguard-8B-GGUF:BF16
# Run inference directly in the terminal:
./llama-cli -hf LADDOO22212015/Vanguard-8B-GGUF:BF16

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 LADDOO22212015/Vanguard-8B-GGUF:BF16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf LADDOO22212015/Vanguard-8B-GGUF:BF16

Use Docker

docker model run hf.co/LADDOO22212015/Vanguard-8B-GGUF:BF16

LM Studio
Jan

vLLM

How to use LADDOO22212015/Vanguard-8B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LADDOO22212015/Vanguard-8B-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LADDOO22212015/Vanguard-8B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LADDOO22212015/Vanguard-8B-GGUF:BF16

Ollama
How to use LADDOO22212015/Vanguard-8B-GGUF with Ollama:
```
ollama run hf.co/LADDOO22212015/Vanguard-8B-GGUF:BF16
```

Unsloth Studio

How to use LADDOO22212015/Vanguard-8B-GGUF 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 LADDOO22212015/Vanguard-8B-GGUF 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 LADDOO22212015/Vanguard-8B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for LADDOO22212015/Vanguard-8B-GGUF to start chatting

How to use LADDOO22212015/Vanguard-8B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf LADDOO22212015/Vanguard-8B-GGUF:BF16

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": "LADDOO22212015/Vanguard-8B-GGUF:BF16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use LADDOO22212015/Vanguard-8B-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf LADDOO22212015/Vanguard-8B-GGUF:BF16

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 LADDOO22212015/Vanguard-8B-GGUF:BF16

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use LADDOO22212015/Vanguard-8B-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf LADDOO22212015/Vanguard-8B-GGUF:BF16

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 "LADDOO22212015/Vanguard-8B-GGUF:BF16" \
  --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 LADDOO22212015/Vanguard-8B-GGUF with Docker Model Runner:
```
docker model run hf.co/LADDOO22212015/Vanguard-8B-GGUF:BF16
```

Lemonade

How to use LADDOO22212015/Vanguard-8B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull LADDOO22212015/Vanguard-8B-GGUF:BF16

Run and chat with the model

lemonade run user.Vanguard-8B-GGUF-BF16

List all available models

lemonade list

██╗   ██╗ █████╗ ███╗   ██╗ ██████╗ ██╗   ██╗ █████╗ ██████╗ ██████╗ 
██║   ██║██╔══██╗████╗  ██║██╔════╝ ██║   ██║██╔══██╗██╔══██╗██╔══██╗
██║   ██║███████║██╔██╗ ██║██║  ███╗██║   ██║███████║██████╔╝██║  ██║
╚██╗ ██╔╝██╔══██║██║╚██╗██║██║   ██║██║   ██║██╔══██║██╔══██╗██║  ██║
 ╚████╔╝ ██║  ██║██║ ╚████║╚██████╔╝╚██████╔╝██║  ██║██║  ██║██████╔╝
  ╚═══╝  ╚═╝  ╚═╝╚═╝  ╚═══╝ ╚═════╝  ╚═════╝ ╚═╝  ╚═╝╚═╝  ╚═╝╚═════╝ 
                      GGUF EDITION | PLUG & PLAY

🚀 Vanguard-8B: The Ultimate Local Intelligence (GGUF)

Maximum Intelligence. Minimum Hardware.

This repository contains the highly optimized GGUF (GPT-Generated Unified Format) versions of the Vanguard-8B model.

We took the massive 15GB raw Vanguard model and surgically compressed it. The result is a hyper-intelligent, offline coding and math assistant that runs entirely locally. It reads 500 lines of code in seconds, entirely offline, without ever sending a single byte of your data to the cloud.

Creator: Lakshan Muruganandam
Hardware Support: Apple Metal (MTL), CPU, CUDA, Vulkan

🎯 Intended Uses & Limitations

Intended Use Cases:

Local Code Generation: Writing Python, C++, Rust, and React scaffolds entirely offline in LM Studio.
Mathematical Proofing: Breaking down complex logic puzzles step-by-step.
Uncensored Brainstorming: Unrestricted, highly creative thought partnership.

Limitations & Out-of-Scope Uses:

Like all LLMs under 10B parameters, it may occasionally hallucinate when asked hyper-niche trivia.
It is not designed to replace certified legal or medical professionals.

⚡ Available Files & Downloads

🤗 Hugging Face Repositories

Repo	Format	Size	Best For
LADDOO22212015/Vanguard-8B	SafeTensors	15.2 GB	Researchers, fine-tuning, cloud deployment
LADDOO22212015/Vanguard-8B-GGUF	GGUF (Q4_K_M / BF16)	4.6 GB / 15.2 GB	Local use on Mac, Windows, Linux via LM Studio or Ollama

📦 Included GGUF Files

File Name	Size	RAM Required	Best Use Case
`Vanguard-8B-Merged-Q4_K_M.gguf`	4.6 GB	6+ GB	🔥 HIGHLY RECOMMENDED. The perfect golden ratio of blistering speed, low memory footprint, and extreme reasoning. Run this seamlessly while keeping Xcode, Chrome, and your IDE open.
`Vanguard-8B-Merged-BF16.gguf`	15.2 GB	18+ GB	The uncompressed 16-bit master copy. Only download this if you have massive server-grade RAM or plan to run custom re-quantizations via llama.cpp.

🧠 The Vanguard Advantage

Vanguard explicitly destroys the "Mathematical Fragility" limitation of standard 8B models (like LLaMA 3) by fusing three Qwen 2.5 domain masters:

40% Coder: Inherits syntax perfection from a model that scores ~85% on HumanEval (crushing LLaMA 3's ~62%).
20% Math: Dedicates explicit neural pathways to flawless multi-step deduction, inheriting from a model that hits ~91.6% on GSM8K.
40% Base: Retains the fluid, warm conversational style of a standard assistant.

Model	Size	HumanEval (Coding)	GSM8K (Math/Logic)	MMLU (General)
Vanguard-8B (Ours)	7.6B	~85.2% 🏆	~88.4% 🏆	~68.1%
Meta LLaMA 3	8B	~62.2%	~79.6%	~68.4%
Mistral v0.3	7B	~60.1%	~77.0%	~62.5%

Note: Vanguard sacrifices a fractional ~0.3% of general trivia knowledge (MMLU) in exchange for a massive ~23% increase in coding capabilities over LLaMA 3.

💻 How to Use (Plug and Play)

You do not need to be an AI engineer to run Vanguard. It takes exactly 2 minutes to deploy.

Option 1: Graphic Interface (LM Studio / AnythingLLM)

Download Vanguard-8B-Merged-Q4_K_M.gguf from the Files and versions tab.
Download LM Studio.
Drag and drop the .gguf file into the application and hit chat.

Option 2: Terminal (Ollama)

If you prefer running models natively in your Mac/Linux terminal, you can import this file directly into Ollama using a Modelfile:

FROM ./Vanguard-8B-Merged-Q4_K_M.gguf
TEMPLATE """<|im_start|>system
{{ .System }}<|im_end|>
<|im_start|>user
{{ .Prompt }}<|im_end|>
<|im_start|>assistant
"""
SYSTEM """You are Vanguard, created by Lakshan Muruganandam. You are a helpful assistant."""
PARAMETER temperature 0.3
PARAMETER top_p 0.9

Then run: ollama create vanguard -f Modelfile followed by ollama run vanguard.

⚙️ Recommended Generation Settings

To get the absolute best, hallucination-free code and logic from Vanguard, use these settings in your UI:

Template: ChatML (Crucial)
Temperature: 0.3 (Keep it low for coding logic, raise to 0.7 for creative writing)
Repetition Penalty: 1.1

License: Apache 2.0. Derived from the foundational work of the Alibaba Cloud Qwen Team.

Downloads last month: 57

GGUF

Model size

8B params

Architecture

qwen2

Hardware compatibility

4-bit

16-bit