Jun Kim, oMLX Creator and Maintainer, Joins Hugging Face to Support the MLX Community
Published September 22, 2026
Hugging Face is excited to welcome Jun Kim โ the creator and maintainer of oMLX โ to the team, where he will focus on supporting and growing the MLX ecosystem.
What This Means for the MLX Community
Apple's MLX framework has become a cornerstone for running machine learning workloads efficiently on Apple Silicon. Jun's oMLX project has been a vital bridge for developers who want to train, fine-tune, and run models natively on Mac hardware โ with an emphasis on simplicity, performance, and community-driven tooling.
By bringing Jun on board, Hugging Face is deepening its commitment to the on-device and local-first ML ecosystem. This includes:
- Continued development of oMLX and related tooling
- Deeper integration between MLX models and the Hugging Face Hub
- Better developer experience for Apple Silicon practitioners
- Community engagement through tutorials, documentation, and events
Why MLX Matters in 2026
As of 2026, running inference and fine-tuning locally has shifted from a niche interest to a mainstream expectation. Apple Silicon Macs โ now spanning M3, M4, and the newer M5 family โ offer substantial unified memory and compute capacity, making them viable machines for serious ML work.
MLX sits at the center of this shift. Its unified memory model, lazy computation, and tight integration with Apple's hardware acceleration make it uniquely suited for local inference of large language models, diffusion models, and multimodal systems.
Projects like oMLX have made it dramatically easier for developers to:
- Convert and quantize models for Apple Silicon
- Run inference with minimal setup
- Fine-tune open models on a laptop
- Share MLX-compatible weights directly on the Hugging Face Hub
Looking Ahead
The Hugging Face team and Jun will be working closely with the MLX community to ship improvements, reduce friction for newcomers, and ensure MLX stays at the forefront of local-first machine learning.
If you're building with MLX โ or curious to start โ now is a great time to get involved. Follow the Hugging Face blog and the oMLX project for updates, and join the growing community of developers bringing state-of-the-art ML to their own hardware.
This article is also available for discussion and updates on GitHub.