What are the core data types in ComfyUI?

The main types you wire between nodes are MODEL (the diffusion UNet), CLIP (the text encoder), VAE (the latent encoder/decoder), CONDITIONING (encoded prompts), LATENT (the working image in latent space), and IMAGE (pixel-space output). A socket only connects to a matching type, which is why the reference lists types explicitly.

Why does KSampler take a LATENT instead of an image?

Stable Diffusion denoises in a compressed latent space, not pixel space. KSampler operates on a LATENT, and you only convert back to a viewable IMAGE at the end with a VAE Decode. Empty Latent Image creates a blank canvas of the right size to start a text-to-image run.

Is this every ComfyUI node?

No. ComfyUI has hundreds of nodes plus thousands from custom node packs. This card covers the core built-in nodes that appear in almost every workflow, which is what most people need when learning or debugging a graph.

Does this run a workflow or just describe nodes?

It is a reference card only — it describes each node's inputs, outputs, and purpose. Build and run actual graphs inside ComfyUI; use this to remember what connects to what.

ComfyUI Node Reference Card

ComfyUI node reference card

ComfyUI is a node-graph interface for Stable Diffusion where each box does one job and you wire them together by data type. The hard part for newcomers is remembering which sockets connect — a MODEL goes here, a LATENT there, a CONDITIONING into the sampler. This searchable card lists the core built-in nodes with their inputs, outputs, and types so you can wire a graph without guesswork.

How it works

Filter by category (loaders, sampling, conditioning, latent, image) or search by name. Each card shows the node’s purpose plus its inputs and outputs with their data types in parentheses. ComfyUI enforces type matching at the socket level, so knowing that KSampler outputs a LATENT and VAE Decode consumes one tells you exactly how they chain.

Tips for building graphs

Follow the data types. A red error usually means a type mismatch — an IMAGE plugged where a LATENT was expected. The type labels here prevent that.
Start from the template. A minimal text-to-image graph is Checkpoint Loader → CLIP Text Encode (×2, positive/negative) → Empty Latent → KSampler → VAE Decode → Save Image.
Decode last. Stay in latent space through sampling and only VAE Decode once at the end; decoding mid-graph wastes compute.
LoRA before sampling. Load LoRA Loader between the checkpoint and the sampler so its MODEL and CLIP edits apply to both conditioning and denoising.