How to chain prompts and models on a visual canvas with Helios Gen

Who it is forA designer producing many variations of a product image.

The problemHosted studios charge per generation and lock workflows in.

The outcomeA reusable visual pipeline for images and video.

What you need first

macOS on Apple silicon for the published build
Node.js to build from source
Model access, such as Codex for some image generation paths

Steps

Download the macOS build from the Releases page
Open Helios Gen and add a prompt node
Connect model and reference image nodes
Run the chain and review outputs

Prompt for your agent

In SegFault42/HeliosGen, build a canvas that takes <reference image> and <prompt> through a model and outputs four variations. Report where outputs were saved.

Caveat. Only the macOS Apple silicon build is published; Windows and Linux need a contributor build.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for SegFault42/HeliosGen and "Chain prompts and models on a visual canvas"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp