Instructions to use TobDeBer/SmartQuant with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TobDeBer/SmartQuant with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="TobDeBer/SmartQuant",
	filename="SmartQuant-Llama-3.3-70B-Instruct.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use TobDeBer/SmartQuant with llama.cpp:

Install from brew

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

Install from WinGet (Windows)

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

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 TobDeBer/SmartQuant
# Run inference directly in the terminal:
./llama-cli -hf TobDeBer/SmartQuant

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 TobDeBer/SmartQuant
# Run inference directly in the terminal:
./build/bin/llama-cli -hf TobDeBer/SmartQuant

Use Docker

docker model run hf.co/TobDeBer/SmartQuant

LM Studio
Jan
Ollama
How to use TobDeBer/SmartQuant with Ollama:
```
ollama run hf.co/TobDeBer/SmartQuant
```

Unsloth Studio new

How to use TobDeBer/SmartQuant 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 TobDeBer/SmartQuant 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 TobDeBer/SmartQuant to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for TobDeBer/SmartQuant to start chatting

Pi new

How to use TobDeBer/SmartQuant with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf TobDeBer/SmartQuant

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": "SmartQuant"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use TobDeBer/SmartQuant with Docker Model Runner:
```
docker model run hf.co/TobDeBer/SmartQuant
```

Lemonade

How to use TobDeBer/SmartQuant with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull TobDeBer/SmartQuant

Run and chat with the model

lemonade run user.SmartQuant-{{QUANT_TAG}}

List all available models

lemonade list

SmartQuant v1 of Llama-3.3-70B-Instruct in just 2.39 bpw.

With just 19.60GB it compares to those two:

Llama-3.3-70B-Instruct-IQ2_XS.gguf IQ2_XS 21.14GB false Low quality, uses SOTA techniques to be usable.
Llama-3.3-70B-Instruct-IQ2_XXS.gguf IQ2_XXS 19.10GB false Very low quality, uses SOTA techniques to be usable.

I'll do some qualification and perplexity runs next.

Downloads last month: 22

GGUF

Model size

8B params

Architecture

granite

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support