Repolode
Submit
huggingface

huggingface/autotrain-advanced

No-code tool from Hugging Face for training and fine-tuning ML models (LLMs, text, image, token tasks) via a web UI, CLI or YAML configs.

View on GitHubWebsite
Stars
4,620
Forks
631
Open issues
4
Last commit
8 days ago

What is it?

AutoTrain Advanced is a no-code tool from Hugging Face for training and fine-tuning machine learning models. It can be driven through a local web UI, a command-line interface, or YAML config files, so users can start a training job without writing training code.

The project covers LLM fine-tuning methods including SFT, ORPO, DPO and reward modeling, along with text classification and regression, token classification, seq2seq, extractive question answering, and image classification and regression. Training can run locally on your own hardware or on Hugging Face Spaces, and models can be pushed to the Hugging Face Hub when finished.

The repository is no longer maintained, so it receives no new features or bug fixes. The maintainers recommend Axolotl, TRL, or transformers.Trainer for ongoing work.

Key features

  • 01Web UI, CLI, and YAML config-based training workflows
  • 02Supports LLM SFT, ORPO, DPO, and reward fine-tuning
  • 03Covers text, token, seq2seq, and image classification or regression tasks
  • 04Can run locally or on Hugging Face Spaces
  • 05Pushes trained models to the Hugging Face Hub

Use cases

  • Fine-tuning a small open LLM on a custom instruction dataset without writing training code
  • Training a text classifier on labeled data through a browser UI
  • Running reproducible training jobs from version-controlled YAML configs

Quick start

  1. 1

    Install the package via pip (requires Python 3.10+)

    pip install autotrain-advanced
  2. 2

    Start the web UI locally

    autotrain app --port 8080 --host 127.0.0.1
  3. 3

    Or train from a YAML config file

    autotrain --config <path_to_config_file>

Things to know

  • The project is no longer maintained, so no new features or bug fixes will be added.
  • Training large models requires substantial compute resources such as a GPU, and PyTorch must be installed separately.

In collections

Natural language processing tools & libraries

Topics

#machine-learning#llm#fine-tuning#no-code#huggingface#nlp#deep-learning#autotrain#natural-language-processing#natural-language-understanding

FAQ

What is autotrain-advanced?
No-code tool from Hugging Face for training and fine-tuning ML models (LLMs, text, image, token tasks) via a web UI, CLI or YAML configs. AutoTrain Advanced is a no-code tool from Hugging Face for training and fine-tuning machine learning models. It can be driven through a local web UI, a command-line interface, or YAML config files, so users can start a training job without writing training code.
Is autotrain-advanced free to use? What license does it have?
autotrain-advanced is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is autotrain-advanced still maintained?
Actively maintained. The last commit to autotrain-advanced was on Oct 29, 2026.
What can I use autotrain-advanced for?
Fine-tuning a small open LLM on a custom instruction dataset without writing training code Training a text classifier on labeled data through a browser UI Running reproducible training jobs from version-controlled YAML configs
How do I install autotrain-advanced?
Install the package via pip (requires Python 3.10+): pip install autotrain-advanced Start the web UI locally: autotrain app --port 8080 --host 127.0.0.1 Or train from a YAML config file: autotrain --config <path_to_config_file>
Where can I get autotrain-advanced?
The source code and installation instructions are on GitHub: https://github.com/huggingface/autotrain-advanced

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects