In simple English
This tool helps you teach an artificial intelligence new information or a specific way of talking. It works with many popular large language models. You do not need to write complex Python code to use it. You can run it through a standard terminal command or use a web page with buttons. It saves you days of setup work when you want to customize your own model.
What it is
A Python framework (a collection of code libraries that give developers a foundation to build on) that unifies fine-tuning for more than one hundred large language models. It provides both a command-line interface (a way to control a program by typing text commands) and a graphical web interface.
What it is good for
Solo builders who want to adapt an open-source language model to understand their own business data or speak in a specific style without writing custom training scripts.
- Supports over one hundred different text and vision models out of the box.
- Includes a built-in web page so you can configure training without writing code.
- Offers memory-saving options like QLoRA to train large models on modest graphics cards.
- Requires an advanced graphics card with several gigabytes of memory to train anything larger than small models.
- Setting up the correct Python environment and dependencies can take patience for non-programmers.
- Training large models takes significant time and electricity.
Writing custom PyTorch training loops from scratch
vLLM
- unslothai/unsloth78k stars
Focuses on fast and memory-efficient fine-tuning specifically for popular open models.
- axolotl-ai-cloud/axolotl13k stars
A popular configuration-driven tool for fine-tuned language models with YAML files.
- mudler/LocalAI50k stars
Serves text, vision, speech, image and video models from one self-hosted service that imitates the paid interfaces.
- janhq/jan45k stars
Downloads and runs language models entirely offline as a desktop application, and opens a local server other programs can use.
- Mintplex-Labs/anything-llm67k stars
One program that reads your documents and answers questions about them, with the model running locally if you want.
git clone --depth 1 https://github.com/hiyouga/LlamaFactory.git cd LlamaFactory pip install -e . pip install -r requirements/metrics.txt