Explore distributed inference and training with MLX

Key Points

• Scale the machine learning workloads across multiple Macs using MLX.
• Explains how to tackle interconnect efficiency, large model inference, request batching, and distributed training challenges.
• Covers how a few Macs on the desk can replace expensive cloud infrastructure for demanding AI workloads.
• The session moves through Distributed communication, Setting up the cluster, Distributed inference and fine-tuning, Model parallelism strategies, Distributed fine-tuning, CLI, Python, Swift, and C++ APIs.
• Key concepts include MLX, RDMA, JACCL, MacBook, LLMs, macOS 26.2.
• Platform coverage: macos.