Building
Nodes & connections
A workflow is a graph of nodes wired together on the canvas. Each node does one job; the edges between them decide what feeds what, and in what order things run.
Reference
Screenshot or asset you upload
Instruction + model
Reads inputs, generates output
Output
Typed React you copy or download
The core node types#
Three node types cover most workflows. You add them from the canvas and connect them by dragging from one node’s output handle to another’s input handle.
Image node
Holds an image — either an input you upload or an output the workflow generated. Input image nodes are how you bring a screenshot, brand asset, or reference onto the canvas. Generated images come back in a chosen aspect ratio: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, or 2:3.
Prompt node
The workhorse. A prompt node holds your instructions, the model to run, and what kind of output to produce. It reads from any nodes wired into it — an image, upstream text, generated code — and produces a new output when it runs. Output types include image, text, SVG, video, and HTML.
Code node
Displays generated code with syntax highlighting — TSX, CSS, JSON, Markdown, SVG, and more. Copy the contents or download the file straight from the node; neither is gated behind a paywall.
Connecting nodes#
Drag from the handle on the right edge of a node (its output) to the handle on the left edge of another (its input). The connection means “feed this node’s result into that one.” A prompt node can take several inputs at once — for example a reference image plus an upstream design brief.
- Outputs flow left-to-right, from source handle to target handle.
- A node can fan out to multiple downstream nodes, and a node can take multiple inputs.
- Cycles aren’t allowed — a node can’t depend on its own output, directly or indirectly.
Reference
Generate component
Extract color palette
How a workflow runs#
When you run a workflow, vmotif sorts the graph into dependency order and executes top-down. A node runs only once everything wired into it has finished, so a prompt waiting on a generated image won’t fire until that image exists. Independent branches run in parallel, and outputs appear on the canvas as each step completes.
Choosing a model#
Each prompt node picks the model that fits its output type. vmotif routes across providers so you can use the strongest model for each job. The in-app model picker is always the source of truth; a sampling of what’s available:
- Images
- Nano Banana Pro (Gemini 3), GPT Image 2, Flux 2 Pro, Flux Kontext, Recraft V4, Imagen 4 Ultra, Grok Imagine.
- Text & code
- Claude Opus 4.7, Claude Sonnet 4.6, GPT-5.2, Gemini 3 Flash, Kimi K2.6, and more.
- SVG
- Gemini 3 Pro and Gemini 3.1 Pro Preview.
- Video
- Veo 3.1, Kling 3.0, Wan 2.6, Seedance 2.0 — text-to-video and image-to-video.