Install Stable Diffusion 3.5 Medium in ComfyUI: A Step-by-Step Guide

Stable Diffusion 3.5 Medium is a powerful model offering impressive image quality with streamlined efficiency. In this guide, we'll walk you through installing Stable Diffusion 3.5 Medium in ComfyUI, a popular node-based UI that simplifies text-to-image generation. Whether you're a beginner or experienced with AI models, this tutorial covers the steps for seamless installation, configuration, and image testing. With ComfyUI, you'll have complete control over generating high-quality images and understanding complex prompts. Let's dive into Stable Diffusion 3.5 Medium's installation and start testing prompts in ComfyUI.

What is Stable Diffusion 3.5 Medium?

Stable Diffusion 3.5 Medium is a smaller, more efficient version of the larger 3.5 model. Despite its size, it delivers high-quality, detailed images, excellent prompt understanding, and optimized resource use, making it ideal for users needing robust performance on mid-range hardware. It is a diffusion Transformer text image generative model and this model generates images based on text prompts.

Set Up ComfyUI for Stable Diffusion

ComfyUI serves as the backbone for managing various Stable Diffusion AI Image generation models, enabling users to effortlessly switch, configure, and test different models in a cohesive, intuitive environment. With its robust setup and community support, ComfyUI provides all the tools necessary to streamline your workflow and enhance your AI model experimentation experience.

What is ComfyUI?

ComfyUI is a versatile, node-based interface for handling model installation and testing. Designed for stability and ease, it's ideal for working with Stable Diffusion models.

Install ComfyUI

Before setting up Stable Diffusion 3.5 Medium, ensure ComfyUI is installed on your system. You can find a straightforward installation guide on ComfyUI's official site or through community tutorials.

How to Install Stable Diffusion 3.5 Medium in ComfyUI

Installing Stable Diffusion 3.5 Medium in ComfyUI involves multiple key steps, each essential to ensure smooth integration and functionality. From downloading the necessary model and clip files to configuring workflow settings, following each step precisely will help you avoid common setup issues and optimize your experience with ComfyUI's powerful, user-friendly interface.

Step 1: Download Model Files

1. Access the Model Card: Visit Stable Diffusion 3.5 Medium model card on Hugging Face.

2. Download Necessary Files: Download the “sd3.5_Medium safetensors” and store them in ComfyUI's models/checkpoint directory.

Step 2: Download Additional Clip Files

ComfyUI requires clip files for text encoding:

Download clip.g, clip.l, and t5xxl_fp16 safetensors from the Hugging Face repository and save them in ComfyUI's models/clip directory.

Step 3: Configure Workflow JSON

Load Workflow: ComfyUI and Stability AI provide a JSON workflow file. Download it and save it into ComfyUI's manager section to enable seamless model operation.

Installation Tips

• File Organization: Keep model and clip files organized for easy access in ComfyUI.

• Compatibility Checks: Ensure your version of ComfyUI is updated to avoid any installation issues by clicking on the Manager and then clicking on the Update All.

Test Image Generation in ComfyUI

Text-to-Image Test

Once Stable Diffusion 3.5 Medium is installed, test it with various prompts to observe its capabilities in detail. Some initial prompts you can try include:

• Simple Scenarios: Nature landscapes, cityscapes, or futuristic scenes.

• Complex Descriptions: Detailed character portraits or dynamic action scenes.

Use Samplers and Schedulers

Experiment with samplers like DPM p2m for smooth outputs or Euler ancestral for sharper details.

Performance Highlights of Stable Diffusion 3.5 Medium

Stable Diffusion 3.5 Medium excels at high-quality rendering and resource efficiency, with notable strengths in:

• Complex Prompt Understanding: It can capture fine-grained details.

• Image Quality: Outputs are vibrant, with clear focus on requested elements.

Also Read: Local Install Guide: FLUX.1 Schnell Model on ComfyUI

Troubleshoot Common Issues

Encountering issues during installation or setup is common, but many problems can be resolved with a few simple checks. Whether it's optimizing VRAM usage for large projects or ensuring files are correctly downloaded and placed, following these troubleshooting steps will help keep your workflow running smoothly and efficiently in ComfyUI.

VRAM Management

When working with high-resolution images, monitor VRAM usage (16GB recommended). Opt for the fp8 model if resources are constrained.

Installation Problems

If files don't load, verify Hugging Face credentials and re-download any corrupted files. Re-check file directories in ComfyUI.

Conclusion

With this comprehensive guide, you're ready to install and test Stable Diffusion 3.5 Medium in ComfyUI, unlocking powerful text-to-image capabilities. Follow these steps, and let ComfyUI bring your prompts to life with impressive visuals and rich detail.