How to Use ComfyUI (Step-by-Step Tutorial)

About this video

Kevin gives a beginner step-by-step tour of ComfyUI, the free, open-source, node-based tool for generating images, video, audio, 3D and more on your own computer. He installs the desktop version (noting the paid cloud version suits people without powerful hardware), selects local running, and loads the Getting Started text-to-image template. ComfyUI itself runs workflows while a separate model creates the image, so he uses Show Missing Models and Download All to install the required model. The workflow loads the checkpoint, takes positive and negative prompts, sets image size and batch count, and samples with settings beginners should leave at default except the seed, which produces variations from the same prompt. Clicking Run generates a chocolate chip cookie image that is auto-saved and viewable in Assets (shortcut A). He then changes the prompt to a double chocolate chip cookie with a melted center and widens the image from 512 by 512 pixels to 768 wide, adds an Image Upscale By node set to 2X between the decode and save steps, and runs it again. He finishes by saving and sharing workflows and recommending the template library — including image to video, image editing, image to 3D and text to song — as starting points to modify rather than building from scratch.

Key takeaways

• Install the free desktop version for local use; ComfyUI runs the workflow but a downloadable model is what actually generates the image.
• Start from the Getting Started text-to-image template, install any missing model it flags, then enter positive and negative prompts before clicking Run.
• Leave sampler settings at their defaults as a beginner, but change the seed to get different variations of the same prompt.
• Workflows are nodes connected in sequence where data flows to the final image; inserting a node such as 2X upscaling between decode and save extends the workflow.
• Save and reuse your workflows, and explore built-in templates for tasks like image to video and image to 3D instead of building everything from scratch.

Episode details: published June 12, 2026 · 9:10 long. Summary written by Self Growth Videos from the episode itself. Business results and any income figures mentioned are Kevin Stratvert’s own claims, shared for education — they are not typical or guaranteed.

← Previous: How to Use Gemini in Google Docs · Next: How to use Claude in Microsoft Word →

All Kevin Stratvert videos · Kevin Stratvert profile · Kevin Stratvert on YouTube