#2 · Primary category: MLOps & Evaluation

LlamaFactory

agent ai deepseek fine-tuning gemma gpt instruction-tuning large-language-models llama llama3 llm lora moe nlp peft qlora quantization qwen rlhf transformers

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Project last updated:08/27/26

GitHub Stars

74.4K

Forks

9.1K

Contributors

296

License

Apache-2.0

Why we included this project

Fine-tuning a modern language model used to mean pulling together trainers and dataset loaders from several different libraries. LLaMA Factory folds that work into one tested workflow that covers more than a hundred open-weight models and vision-language variants, with LoRA, QLoRA, and full-parameter training available from the same interface. Teams adapting a base model to their own data, whether for instruction tuning, preference alignment, or continued pretraining, can go from raw data to a tuned checkpoint without rebuilding the plumbing each time. It ships a web UI, a command-line tool, and a Python API, so it suits both researchers poking at experiments interactively and engineers who want to drop fine-tuning into an automated pipeline. For anyone weighing how to productionize model customization, it is a solid reference for what a mature, community-tested fine-tuning stack looks like.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category