Understanding, one node at a time.
Hands-on lessons from the app’s current catalog. Build on the real canvas and get a reason for each step.
76 lessons found
ComfyUI basics3
How ComfyUI works
You learn what a graph, the queue and the models folder are, and find a node, an output and a model file yourself.
Reading someone else's workflow
You learn to read groups, notes, reroutes, primitives, Set and Get, bypassed nodes and subgraphs, and give two fields one value.
When the graph does not fit the graphics card
You learn why a graph runs out of memory and what to change, and set the picture graph up to need less of it.
Pictures: first steps4
A first picture from text
You build a graph of seven nodes that draws a picture from a prompt.
The prompt: how a model reads it
You learn what a model does with text and how to steer a picture with words.
The sampler's settings
You set the fields of the sampler the way ComfyUI's own SDXL template sets them, and learn what each one does.
Refining with a second model
A second model redraws a finished picture: the composition comes from the first, the texture and style from the second.
Video14
Video from text
You build a Wan 2.2 graph that makes a five-second clip from a prompt.
A prompt for video
You learn to describe a clip so that the model understands the motion and the camera.
Video from a picture
You bring a finished picture to life: it becomes the first frame of the clip.
LoRA: add a style or a subject
You connect a LoRA to both models and put its words into the prompt.
Fast video in four steps
You cut twenty steps down to four with a speed-up LoRA.
Two models in one graph
A picture model draws the first frame and Wan brings it to life. Each has its own LoRAs and its own prompt.
Video on a modest graphics card: Wan 2.2 5B
You build a one-model Wan 2.2 graph that makes a five-second 720p clip from a prompt, or from an image.
When a video does not fit in memory
You change the graph of the small Wan model so that it needs less graphics memory, and you know what each change costs.
Video guided by another video: Wan 2.2 Fun Control
Your image comes to life with the motion of another video: its depth, its pose or its edges.
A portrait that speaks: HuMo
A person from an image says the words of a sound file, with ComfyUI's own nodes, no pack.
Video with sound: LTX-2.5
One run draws a clip and its sound together, from a prompt.
Video from text: HunyuanVideo 1.5
You build the HunyuanVideo 1.5 graph with a custom sampler and get a five-second 720p clip.
Video from text: Kandinsky 5 Lite
You build a light one-model graph that makes a five-second clip from a prompt.
Smoother motion: more frames between frames
A finished clip gets twice as many frames and plays smoothly at twice the frame rate.
WanVideoWrapper10
Video on WanVideoWrapper
You build a Wan 2.2 graph on the nodes of the WanVideoWrapper pack and see how it differs from the built-in one.
Video from a picture on WanVideoWrapper
A clip made on WanVideoWrapper nodes starts from your picture.
VACE: motion from another video
A clip on WanVideoWrapper repeats the motion, the depth or the outlines of another video.
Animation by pose: Wan Animate
The character from your picture repeats the moves and the face of the person in a video.
Animation by pose: UniAnimate
The character from a picture repeats the moves of a person in a video. A Wan 2.1 model and one LoRA are enough.
A talking video: InfiniteTalk
A portrait says the words of a sound file, the lips in time with them.
Two people talking: InfiniteTalk
Two people in one picture speak, each with their own voice.
A talking portrait: FantasyTalking
A portrait says a short voice recording. The graph is simpler than InfiniteTalk's, the clip shorter.
A video with sound: Ovi
One run draws the clip and its sound at once: the character's speech, steps, street noise.
Another camera: ReCamMaster
A finished video is drawn again as if the camera had moved another way.
Pictures: models10
SDXL done properly
You set an SDXL graph to the size and the settings the family was trained for.
Pony and Illustrious: tags instead of sentences
You write a tag prompt with the quality words of the model's page and set the size and sampler such models are used with.
Choosing a sampler and a scheduler
You put two samplers side by side that differ in one setting only, and learn to choose by comparing.
Which loader for which file
You find the right loader for four kinds of model files and learn where ComfyUI looks for each.
Flux.1: a picture from separate files
You build a Flux graph of ten nodes, where the model, the two text encoders and the VAE come from loaders of their own.
Z-Image Turbo: a picture in eight steps
You build a graph for a distilled model that draws in 8 steps with cfg 1, and learn why its negative prompt does nothing.
Qwen-Image: words inside the picture
You build a Qwen-Image graph with a real negative prompt, the shift of the family and its own size.
Flux.2 Klein: a sampler built from parts
You build the graph of a small Flux.2 model, where the sampler is put together from five nodes.
SD 3.5: one file and a new empty latent
You build an SD 3.5 graph from one checkpoint with the empty latent and the settings of this family.
A quantised model: GGUF
You put a GGUF model into a graph of separate files, so a large model fits a smaller video card.
Video: the anchor picture4
The start picture is the anchor
Your picture becomes the first frame of the clip, and you know how to trade motion for likeness.
Two anchors: first and last frame
You give two images, and the model draws the clip that goes from the first to the last.
A reference for the whole clip: Wan VACE
A person or an object from your picture stays itself in every frame, not only in the first.
A longer video in parts
Two parts of a clip join through a shared anchor frame into one longer video.
Packs and tools8
Load, cut and save video: VideoHelperSuite
You load a piece of a video at a chosen frame rate and save it with its sound.
A larger, sharper video: SeedVR2
A low-resolution clip is redrawn at 1080p with its details restored.
Remove an object from a video: ProPainter
An object marked by a mask video disappears, and the background behind it is filled in from the neighbouring frames.
ControlNet from a depth map
The comfyui_controlnet_aux pack turns a picture into a depth map, and the model draws a new scene in the same space.
A reference picture with IP-Adapter
An SD 1.5 or SDXL model takes the style of a sample picture through the ComfyUI_IPAdapter_plus pack.
Fixing faces and hands with a detailer
The Impact Pack finds the faces in a finished picture and draws each one again at a larger size, so small faces come out sharp.
Two versions of a prompt with one switch
You keep two prompts in one graph and choose between them with one yes/no value, the way ComfyUI's templates do.
A language model writes the prompt
You let a language model inside ComfyUI turn a short idea into a detailed prompt, and compare it with your own by the switch.
Pictures: techniques12
A picture from a picture
A picture you already have becomes a new one: the same composition, drawn again by the model as the prompt says.
Inpainting: redraw part of a picture
You paint a mask over part of a picture, and the model draws only that part again, by a new prompt.
Outpainting: a picture beyond its edges
A picture gets a new strip on one side, and the model draws what lies beyond the frame.
A mask by words
ComfyUI finds an object by its name and gives its mask to the inpainting node, with no painting by hand.
Upscaling with a model
The finished picture is enlarged by a model that draws sharp detail instead of stretching the pixels.
A second pass after upscaling (hires fix)
The enlarged picture is drawn once more by the model: sharp detail at a size the model cannot draw in one go.
ControlNet: drawing along the edges of a picture
The model draws a new picture that keeps the outlines of a sample picture.
LoRA for pictures
Two LoRAs in a chain change the style and the subject of a picture model, each with its own strength.
LoRA for a character: face, body, clothes, pose
Three LoRAs on one picture, each tuned on its own, the face kept while a body or clothing LoRA works.
Editing a picture by words
A model made for editing changes a picture as an instruction says: "make the sofa green", "remove the cup".
A reference picture for style (Flux Redux)
A Flux.1 picture takes its look from a sample picture: colours, materials, mood.
Removing a background
A picture whose background is made transparent, saved as PNG.
Sound and speech5
Music from text: ACE-Step 1.5
You build a graph that writes a song from a list of styles and your lyrics.
A sound from text: Stable Audio
You build a graph that makes a sound effect or a short piece of music from a prompt.
Speech from text: Chatterbox
You build a graph that reads your text aloud in the voice of a short recording.
Sound for a silent video: MMAudio
You build a graph that listens to the frames of a silent clip and gives it a sound in time with them.
Speech to text: Whisper
You build a graph that writes down what is said in a recording and saves subtitles.
3D1
A picture into a 3D model: Hunyuan3D 2
You build a graph that turns one picture of an object into a 3D model you can open in other programs.
Faces and characters5
The same character in a new picture: FLUX.1 Kontext
You build a graph that draws your invented character again, in a new place or pose, with the same face and clothes.
One character in many pictures
You learn what holds a character's identity and practise a new pose and new clothes while the face stays.
The same face from one photo: IP-Adapter FaceID
You add IP-Adapter FaceID to the first picture graph so that a photo of a face drives the face of the new picture.
Put a face onto a picture: ReActor
You build a graph that puts a face from one picture onto a person in another picture you already have.
Restore a face in an old picture: ReActor
You build a graph that redraws blurred or damaged faces in a picture with a face restoration model.
Some lessons share the topic index in the wiki. The complete steps are built into the desktop app.
Your next connection starts here.
A free desktop tutor. A real canvas. A reason for every step.
Download for LinuxLinux x64 · Free