📹 ComfyUI 101

10 min

Learn to build AI workflows from a blank canvas. 10 lessons, zero assumptions.

https://www.youtube.com/embed/1CePXVRBm60

What this course covers

This course walks you through ComfyUI from the ground up. You'll learn how the interface works, what nodes and connections actually do, and how to build your own workflows from scratch.

By the end, you'll know how to generate images from text, transform existing images, upscale to 4K, and control composition with ControlNet. You'll also understand the models behind all of it: diffusion models, LoRA, CLIP, and VAE.

No prior experience assumed. If you've never opened a node-based tool, that's fine. If you have some experience and want to fill gaps, you'll get that too.

Every workflow in this course runs directly in Floyo. No local install needed. Open your browser and follow along.

Start the course

Ep 1: What is ComfyUI?Ep1: What is ComfyUI?

Nodes, workflows, and the blank canvas

Ep 2: Run your first ComfyUI workflowEp 3: Run your first ComfyUI workflow

Load a template, set a prompt, generate

Ep 3: Nodes 101Ep 3

The color-coded connector system

Ep 4: Text-to-Image workflowsYour first Text-to-Image workflow

Build a complete workflow node by node

Ep 5: KSamplerEp 5:

Seed, steps, CFG, sampler, denoise

Ep 6: SubgraphsEp 6

Group nodes into clean, reusable blocks

Ep 7: ModelsEp 7: Models

Diffusion, LoRA, CLIP, VAE, ControlNet

Ep 8: Image-to-Image workflowsEp 8: Image-to-Image workflows

Transform photos and sketches with AI

Ep 9: UpscalingEp 9:

Two methods to get sharp results at 4K

Ep 10: ControlNet basicsEp 10

Canny, depth, OpenPose, and line ar

Who this course is for

Designers, photographers, content creators, developers, or anyone curious about AI image generation. The only prerequisite is a browser.

This is not a prompt engineering course. You won't learn how to write better prompts. You'll learn how ComfyUI itself works so you can build, modify, and understand any workflow you come across.

The course is taught by Sebastian Kamph and hosted on Floyo.

Frequently Asked

Start with Episode 1. You'll load a template, set a prompt, and generate your first image in under two minutes. No setup needed if you're using Floyo. Episodes 1 and 3 give you the concepts behind what's happening, and Episode 4 teaches you to build a workflow from scratch.

No. Everything runs on Floyo in the browser. You don't need to install ComfyUI locally or have specific hardware.

Midjourney and DALL-E give you a prompt box and a result. ComfyUI gives you the full pipeline. You choose the model, control the sampler, decide how the image gets processed at every step. More control, more flexibility, steeper initial learning curve. That's what this course is for.

Yes. ComfyUI is open source. The concepts, nodes, and workflows taught here work the same way whether you're running on Floyo or locally on your own machine.

Stable Diffusion is an AI model. ComfyUI is a tool that runs AI models, including Stable Diffusion. But ComfyUI also runs Flux, WAN, ZImagine, and others. This course teaches ComfyUI the tool, not any single model.

Each episode is a short video lesson. You can get through the full course in an afternoon, or take it one episode at a time over a week. Go at whatever pace works.

📹 ComfyUI 101

﻿

Start the course

Who this course is for

Frequently Asked