TechPrufer
Get started
Replicate
Replicate logoReplicate · Feature trail

Local ML Model Deployment

1 post · Aug 2022 – Jun 2024

What this trail covers

Replicate enables users to run advanced ML models locally on their own hardware, starting with Stable Diffusion on Apple Silicon Macs. This includes providing detailed instructions for cloning repositories, setting up environments, and running models, as well as exploring alternative CLI tools like DiffusionKit for easier deployment. The focus is on empowering users with local inference capabilities for powerful generative AI models. This post extends this by providing instructions for running Stable Diffusion 3 locally using ComfyUI and Cog.

All Replicate trails
  1. Run Stable Diffusion 3 on your Apple Silicon MacThis post details how to run Stable Diffusion 3 locally on Apple Silicon Macs. It provides two methods: a direct Python script using the MPS backend and the DiffusionKit CLI tool with Python MLX. It includes prerequisites (Mac with M-series chip, Git), steps for cloning the repository, setting up virtual environments, installing dependencies, and running the generation script. It also outlines customization options for seed, prompt, dimensions, inference steps, and guidance scale. A speed comparison between the two methods on an M3 Max MacBook Pro is provided, highlighting the performance differences with various DiffusionKit flags like `--a16`, `--w16`, and `--t5`.
  2. Run Stable Diffusion 3 on your own machine with ComfyUIThis post provides a step-by-step guide for users to run Stable Diffusion 3 locally on their own GPU-equipped machines using ComfyUI. It details the prerequisites, installation of Cog, cloning the ComfyUI repository, starting the ComfyUI server, and loading the SD3 workflow. It also explains how to generate an image with a text prompt.
  3. Generate images in one second on your Mac using a latent consistency modelThis post provides detailed instructions for setting up and running Latent Consistency Models (LCMs) locally on M1/M2 Macs. It covers prerequisites (hardware, OS, Python), Python setup via Homebrew, cloning the repository, setting up a virtual environment, installing dependencies, and running the model with specific parameters. It also includes performance benchmarks based on Mac hardware and encourages users to share their results.
  4. Run Stable Diffusion on your M1 Mac’s GPUThis post provides a detailed guide for running Stable Diffusion locally on M1/M2 Macs, focusing on using pip and binary wheels for easier dependency management. It outlines prerequisites, Python 3.10 setup via Homebrew, cloning a specific fork of Stable Diffusion, installing dependencies with `pip install -r requirements.txt`, downloading model weights, and executing the `txt2img.py` script. It also mentions alternative next steps like using Replicate's API or exploring other forks like Deforum.