Skip to content

Quick Start Guide

Get started with ModelForge in just a few minutes!

Prerequisites

Before you begin, ensure you have:

  • Python 3.11.x (Python 3.12 is not yet supported)
  • NVIDIA GPU with at least 4GB VRAM (6GB+ recommended)
  • CUDA installed and configured
  • HuggingFace Account with an access token (Create one here)

Windows Users: See the Windows Installation Guide for important platform-specific instructions, especially if you want to use the Unsloth provider.

Installation

1. Install ModelForge

pip install modelforge-finetuning

2. Install PyTorch with CUDA Support

Visit the PyTorch installation page and select your CUDA version.

For example, for CUDA 12.6: 

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126

For CUDA 11.8: 

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

3. Set Your HuggingFace Token

Linux: 

export HUGGINGFACE_TOKEN=your_huggingface_token

Windows PowerShell: 

$env:HUGGINGFACE_TOKEN="your_huggingface_token"

Windows CMD: 

set HUGGINGFACE_TOKEN=your_huggingface_token

Or use a .env file (all platforms): 

echo "HUGGINGFACE_TOKEN=your_huggingface_token" > .env

Launch ModelForge

modelforge run

Open your browser and navigate to: 

http://localhost:8000

That's it! You should see the ModelForge web interface.

Your First Fine-Tuning Job

1. Prepare Your Dataset

Create a JSONL file with your training data. The format depends on your task:

Text Generation: 

{"input": "What is machine learning?", "output": "Machine learning is a subset of AI..."}
{"input": "Explain neural networks", "output": "Neural networks are computing systems..."}

Summarization: 

{"input": "Long article text here...", "output": "Short summary here."}

Question Answering: 

{"context": "Context paragraph...", "question": "What is X?", "answer": "X is..."}

See Dataset Formats for detailed information.

2. Upload Your Dataset

  1. In the ModelForge UI, click "Upload Dataset"
  2. Select your JSONL file
  3. Wait for validation to complete

3. Configure Training

Select your training options:

  • Task: Choose text-generation, summarization, or extractive-question-answering
  • Model: Pick a recommended model for your GPU (or browse for others)
  • Provider: Choose huggingface (standard) or unsloth (2x faster, requires WSL/Docker on Windows)
  • Strategy: Choose sft (standard) or qlora (memory-efficient)
  • Hardware Profile: Select based on your GPU VRAM

4. Start Training

Click "Start Training" and monitor progress in real-time!

5. Test Your Model

Once training completes: 1. Navigate to the Playground tab 2. Select your fine-tuned model 3. Enter a prompt and see the results!

Next Steps

Stopping ModelForge

To stop the application and free up resources:

# Press Ctrl+C in the terminal running ModelForge

Running ModelForge Again

Simply run:

modelforge run

Your previous models and settings are preserved!

Need Help?

Congratulations! You're ready to start fine-tuning LLMs! 🎉