Synth Node – Build Your Own AI Image & Video Pipeline, Node by Node

Synth Node lets you design your own AI image and video pipeline as a visual node-based workflow.

Instead of being locked into one fixed workflow, you choose exactly the functions you need and connect the nodes into your own pipeline.

For example:

Input image → keep character → edit image → upscale → generate video → preview

or:

Prompt → generate image → upscale → generate video

Each node handles one function, so it's easy to change, extend or reuse the workflow for different purposes.

Core nodes

Image Source

Brings an input image into the workflow — for editing an existing image, keeping a character from a source image, turning an image into video, upscaling, or using an image as a reference for later steps.

Subject Lock

Keeps a character or subject consistent throughout the workflow — useful when generating many images or videos while keeping recognisable features.

Text To Image

Generates a new image from a text prompt. Enter your description, then connect this node to the next steps, e.g. Text To Image → Upscale Image → Image To Video.

Image To Image

Edits or transforms an image based on an input image — change style, change background, swap outfit, transform the concept, or create a new version from a source image.

Upscale Image

Increases resolution and improves image quality before use or before turning it into video — especially useful for low-resolution inputs.

Image To Video

Turns an image into AI video. Place it at the end of the pipeline after the image has been generated, edited or upscaled, e.g. Image Source → Image To Image → Upscale → Image To Video.

Preview

View the workflow's final result. Connect a Preview node to any step to check the output before continuing to edit the workflow.