Apple has enhanced its machine learning offerings with the re-release of MLX integrated with Swift, allowing for local execution of Large Language Models (LLMs). Following this, the company introduced MLXServer, a new project aimed at simplifying the creation of APIs for these models. MLXServer, developed by Mustafa (@maxaljadery) and Siddharth, provides an easy setup for running open-source models optimized for Apple's Metal. It includes features like text generation, chat functionalities, and converting models, accessible via HTTP endpoints. The tool, described as a Python endpoint, is accessible via a simple pip install command, indicating Apple's commitment to making machine learning more accessible to developers. This move positions Apple as a strong competitor against TensorFlow and PyTorch in the machine learning space.
As part of further concessions to the EU DMA, Apple is now saying that developers don't need to fire up their own App Stores, and instead, they can sell apps directly from their websites. By @WGallagher https://t.co/7VwdhH9UXL
Wanna learn how to deploy ML models to production? Let's go step-by-step, with a real-world example 🧵↓ https://t.co/K7es1F4wqK
Apple is going all out with MLX. a few days ago they rereleased MLX with Swift so you can run LLMs locally. now they’re onto MLXServer so you can build APIs around them more easily. solid TF/Pytorch competitor in the making. https://t.co/8auXfSYvax
Exciting new project: MLXServer An easy way to get started with LLMs locally. HTTP endpoints for text generation, chat, converting models, and more. Setup: pip install mlxserver Docs: https://t.co/mLCWxUdcec Example: https://t.co/DEQLHOSAZp https://t.co/zSMgfoIGz1
Mustafa (@maxaljadery) and I are excited to announce MLXserver: a Python endpoint for downloading and performing inference with open-source models optimized for Apple metal ⚙️ Docs: https://t.co/69nBje4BJk https://t.co/vnLtMSJYtL
Learn how to use MAX Engine 🏎️ and MAX Serving⚡️ to optimize and deploy🚀 AI models. Using a simple computer vision example, we walk you through end-to-end cloud model hosting workflow and invocation using client side APIs. Jupyter notebook included! 🎉 ⬇️ https://t.co/xpkK1sEppD