LlamaFactory vs torchtune
LLM fine-tuning workflows compared
LlamaFactory and torchtune both provide Python-based workflows for fine-tuning and post-training language models. LlamaFactory also targets vision models and offers CLI and web interfaces, while torchtune centers on editable PyTorch recipes and is no longer actively maintained.
LlamaFactory: Fine-Tune Large Language and Vision Models
LlamaFactory provides CLI and web interfaces for fine-tuning a broad range of language and vision models. It supports parameter-efficient methods and preference training, with workflows for training, inference, and model export.
torchtune: Fine-Tune and Post-Train Large Language Models
torchtune is a PyTorch library for configuring and running LLM post-training workflows, from supervised fine-tuning to preference optimization. It is aimed at developers who want editable recipes and model-specific configs, but is no longer actively maintained.
| LlamaFactory | torchtune | |
|---|---|---|
| Language | Python | Python |
| License | Apache-2.0 | BSD-3-Clause |
| Stars | 75.3k | 5.8k |
| Forks | 9.2k | 756 |
| Last analyzed | Oct 3, 2026 | Oct 3, 2026 |
Key differences
- LlamaFactory covers language and vision workflows, while torchtune focuses on large language models.
- LlamaFactory offers CLI commands and a Gradio web interface; torchtune organizes workflows around editable recipes and YAML configurations.
- LlamaFactory includes inference and model export workflows, while torchtune emphasizes post-training recipes and evaluation workflows.
- LlamaFactory is licensed under Apache-2.0; torchtune is licensed under BSD-3-Clause.
- LlamaFactory is not archived, while torchtune's development wound down in 2025 and it is no longer actively maintained.
- LlamaFactory supports a broad set of training methods and model families; torchtune provides configurable PyTorch implementations for a defined set of models and methods.
Choose LlamaFactory if you…
- need CLI and web interfaces for model adaptation.
- want workflows spanning language and supported vision, video, or audio tasks.
- need training, inference, and model export in one framework.
Choose torchtune if you…
- want to inspect and edit PyTorch recipes and YAML configurations.
- need recipes for methods such as knowledge distillation, DPO, PPO, or GRPO.
- are maintaining or studying an existing torchtune workflow and can plan around limited upstream support.
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.