Get in Touch

Course Outline

Introduction to ComfyUI and Visual AI Content Creation

  • Understanding ComfyUI and the current visual AI landscape
  • Differences between node-based workflows and traditional creative tools
  • Supported media types: image, video, 3D, and audio

Installation, Setup, and First Generation

  • Using ComfyUI Desktop for Windows and macOS
  • Overview of manual installation options and GPU support
  • Executing a first image generation workflow

The Node Graph Interface and Core Concepts

  • Navigating the canvas, zooming, and selecting nodes
  • Understanding nodes, links, properties, and dependencies
  • The queue system, execution order, and partial re-execution

Core Nodes: Loaders, Samplers, Conditioning, and Outputs

  • Checkpoint loaders, CLIP loaders, and VAE loaders
  • Samplers, schedulers, and key generation parameters
  • Conditioning techniques using positive and negative prompts

Working with Models: Checkpoints, LoRAs, VAEs, and Embeddings

  • Model types and file formats, including safetensors and ckpt
  • Utilizing LoRAs for style and character control
  • Employing embeddings and textual inversion

Controlled Generation: ControlNet, IP-Adapter, and Inpainting

  • Using ControlNet for pose, depth, and edge-guided outputs
  • Leveraging IP-Adapter for image-based style references
  • Techniques for inpainting and outpainting

Image Refinement: Upscaling, Compositing, and Area Composition

  • Upscale models such as ESRGAN, SwinIR, and their variants
  • Workflows for high-resolution fixes
  • Area composition for creating multi-region images

Video Generation Workflows

  • Supported video models: Wan, Hunyuan Video, Mochi, and LTX-Video
  • Frame-by-frame generation and interpolation methods
  • Pipelines for image-to-video and text-to-video conversion

Custom Nodes and the Community Ecosystem

  • Navigating the ComfyUI Manager and Registry
  • Finding, installing, and evaluating custom nodes
  • Accessing community workflows via Comfy Workflows

Workflow Management, Optimization, and Sharing

  • Saving and loading workflows as JSON files
  • Embedding workflow data directly into PNG and WebP files
  • Strategies for memory management, batching, and VRAM optimization

App Mode, API, and Production Pipelines

  • Creating simplified user interfaces with App Mode
  • Exposing workflows as accessible API endpoints
  • Deployment options using Comfy Cloud and Comfy Enterprise

Troubleshooting, Performance, and Best Practices

  • Addressing common errors and applying debugging strategies
  • Smart memory offloading and operating with low VRAM
  • Configuring model organization and search paths

Requirements

  • Fundamental computer literacy and familiarity with file systems
  • No prior experience in AI or programming is required

Audience

  • Digital artists and visual content creators
  • Designers and creative industry professionals
  • AI practitioners interested in exploring visual generation tools
  • Technical artists and specialists in production pipelines
 14 Hours

Number of participants


Price per participant

Testimonials (1)

Upcoming Courses

Related Categories