Image Node & Generation Models
Bring images into your workflows — generate with AI, upload, or fetch from a URL.
The Image node holds an image that other nodes can reference and process. Generate one with AI, drag one in, or pull one from the web, then feed it to vision, image-processing, or design steps.
Generation models
Generate images from a text prompt with multiple AI models:
- DALL·E 2 & 3 (OpenAI)
- Nano Banana (Google's fast model)
- Stable Diffusion (Hugging Face, Stability AI)
- Local models via your own endpoints
Pick the model that fits your need — speed, style, or running fully on your own machine.
Bring images in
- Generate with AI from a prompt.
- Upload by dragging a file onto the node.
- From a URL to fetch an external image.
Reference by name
Other nodes can reference an image by its node name, so paths resolve automatically — no copy-pasting data around. For example, ask a web or design step to "use 'HeroImage' as the background."
Feed it downstream
- Run computer-vision operations — OCR, describe, classify. See Vision.
- Transform it — resize, crop, convert, edge-detect, generate icon sets. See Image Operations.
- Trace it into editable vectors. See ToVec.