This repo contains a variety of standalone examples using the MLX framework.
The MNIST example is a good starting point to learn how to use MLX.
Some more useful examples are listed below.
- MLX LM a package for LLM text generation, fine-tuning, and more.
- Transformer language model training.
- Minimal examples of large scale text generation with LLaMA, Mistral, and more in the LLMs directory.
- A mixture-of-experts (MoE) language model with Mixtral 8x7B.
- Parameter efficient fine-tuning with LoRA or QLoRA.
- Text-to-text multi-task Transformers with T5.
- Bidirectional language understanding with BERT.
- Generating images
- Image classification using ResNets on CIFAR-10.
- Convolutional variational autoencoder (CVAE) on MNIST.
- Speech recognition with OpenAI's Whisper.
- Audio compression and generation with Meta's EnCodec.
- Joint text and image embeddings with CLIP.
- Text generation from image and text inputs with LLaVA.
- Image segmentation with Segment Anything (SAM).
- Semi-supervised learning on graph-structured data with GCN.
- Real NVP normalizing flow for density estimation and sampling.
Note: You can now directly download a few converted checkpoints from the MLX Community organization on Hugging Face. We encourage you to join the community and contribute new models.
We are grateful for all of our contributors. If you contribute to MLX Examples and wish to be acknowledged, please add your name to the list in your pull request.
The MLX software suite was initially developed with equal contribution by Awni Hannun, Jagrit Digani, Angelos Katharopoulos, and Ronan Collobert. If you find MLX Examples useful in your research and wish to cite it, please use the following BibTex entry:
@software{mlx2023,
author = {Awni Hannun and Jagrit Digani and Angelos Katharopoulos and Ronan Collobert},
title = {{MLX}: Efficient and flexible machine learning on Apple silicon},
url = {https://github.com/ml-explore},
version = {0.0},
year = {2023},
}
