🔍 Read the full analysis: oMLX Creator And Maintainer Jun Kim Joins Hugging Face on ThorstenMeyerAI.com
Get monitors, keyboards and dev gear delivered free — and shop member deals
- Fast, free delivery on millions of items
- Access to Prime Big Deal Days deals on October 6–7
- Prime Video, Amazon Music and more included
TL;DR
Jun Kim, creator and maintainer of the open-source oMLX inference engine, has joined Hugging Face to work on Apple’s MLX ecosystem. The company says oMLX will receive funded maintenance while remaining Apache 2.0 licensed under Kim’s leadership; timelines and funding details were not disclosed.
Hugging Face said the move is intended to give oMLX greater stability and allow faster development. The company described the project as moving from a side job to a fully maintained and funded effort, which it said would help Kim guide contributors and plan for the long term. The announcement did not specify the level or structure of that funding.
Kim will focus on Apple’s MLX ecosystem, including work to make it easier to move from a model definition in the Transformers library to a reference MLX implementation that multiple inference engines can use. Hugging Face said this could make it easier for new Transformers models to run on MLX. It did not provide a delivery schedule or name specific models or features.
The company said it has supported MLX since its release in late 2023 and hosts MLX models on its Hub. It also cited existing work with projects including mlx-lm, mlx-vlm and LM Studio, and said it hopes to strengthen collaboration with teams across the ecosystem. Those comments describe its intentions; the announcement did not commit to particular joint releases.
More Support for Mac-Based AI
oMLX’s move into a funded role could give its users and contributors a more dependable path for running models on Apple Silicon. A project with dedicated maintenance may be better positioned to address bugs, review contributions and keep pace with changes in the surrounding MLX software. Those are expected benefits described by Hugging Face, not results already demonstrated by the announcement.
The appointment also signals Hugging Face’s investment in local AI, where models run on a user’s own hardware rather than relying on a cloud service. Better links between Transformers model definitions and MLX implementations could help developers bring more models to Apple devices. Whether that work benefits the wider MLX ecosystem will depend on the contributions and collaborations that follow.
How oMLX Fits the MLX Stack
MLX is Apple’s machine learning framework designed for Apple Silicon. Released in late 2023, it provides a foundation for running and training models on Apple hardware. Projects such as mlx-lm and mlx-vlm build tools and model support on top of that framework.
oMLX is an open-source inference engine in this ecosystem, created and maintained by Kim. Hugging Face’s Hub provides a place for users to find and publish MLX-format models. The company said it has supported MLX-related work since the framework’s release and described the new role as part of that ongoing effort.
““We are completely invested in local AI, and MLX is a central piece of the ecosystem.””
— Hugging Face, in its announcement
Funding Terms and Timelines Remain Open
Hugging Face did not state Kim’s job title, team placement or start date, or disclose the financial terms or scope of oMLX’s funding. It also gave no timeline for the Transformers-to-MLX work or named specific planned releases.
The company said it would like to upstream work where appropriate and strengthen ties with other MLX projects. The announcement does not establish which contributions will be shared, when that may happen, or how the project’s priorities will be set in practice. The effect of the new arrangement on oMLX’s release pace and the wider MLX stack remains to be seen.
Future oMLX releases and repository updates should show how the new funded maintenance affects stability and development pace. Progress on a path from Transformers model definitions to MLX implementations will also indicate whether Hugging Face’s stated technical focus is taking shape.
Further announcements or contributions involving mlx-lm, mlx-vlm, LM Studio and other MLX teams may clarify how broad the collaboration becomes. Hugging Face said it will continue supporting MLX models on the Hub, but it has not set a schedule for these efforts.
Key Questions
What is oMLX?
oMLX is an open-source inference engine for Apple’s MLX ecosystem. Jun Kim created and maintained the project.
What changes for oMLX under Hugging Face?
Hugging Face says oMLX will become a funded, fully maintained project, with Kim continuing to lead it. The company did not publish funding terms or a schedule for changes.
Will oMLX remain open source?
According to Hugging Face’s announcement, oMLX will remain under the Apache 2.0 license.
What technical work did Hugging Face describe?
The company said it wants to streamline the path from Transformers model definitions to reference MLX implementations that different inference engines can use. It has not announced a timeline or specific deliverables.
Primary source: Hugging Face · via ThorstenMeyerAI.com
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
