A local node-based canvas for AI image, video and storyboard creation, with an in-canvas agent, a 3D director studio and pluggable generation providers.
SegFault42/HeliosGen
A self-hosted desktop app for building AI image and video generation pipelines visually on an infinite node-based canvas, using your own kie.ai key.
Open-source alternative to:Higgsfield
OpenArt
Freepik
- Stars
- 3,377
- Forks
- 373
- Open issues
- 13
- Last commit
- 2 months ago
What is it?
HeliosGen is a desktop app for chaining AI image and video generation steps into reusable workflows. You connect prompts, models, reference images and automation steps as nodes on an infinite canvas, and the pipeline can run in parallel or sequentially.
The app runs locally: generations, uploads, folders, workflows and settings are stored in a local SQLite database and media is saved to a local folder. Generation requests go through kie.ai using your own API key, and credits are purchased directly on that account. Supported models include GPT Image 2, Nano Banana variants, Seedream, Z-Image and Grok Imagine for images, and Veo 3.1, Kling 3.0, Seedance 2.0 and others for video.
The desktop shell is built with Tauri 2 and the UI with Next.js and React. A macOS build is published on the releases page, while Windows and Linux currently need to be built from source. For GPT Image 2 it can optionally generate through a local Codex CLI installation instead of kie.ai credits.
Key features
- 01Infinite node-based canvas for building generation pipelines
- 02Image and video generation with multiple models
- 03Reference image support in workflows
- 04Parallel and sequential pipeline execution
- 05Local SQLite storage with media saved to disk
- 06Optional GPT Image 2 generation through Codex CLI
Use cases
- Producing a batch of images or videos from one prompt across several models
- Reusing a multi-step generation workflow with different reference images
- Keeping generated media and prompt history on a local machine
Quick start
- 1
Download the macOS build from the Releases page
- 2
Open Settings → API Keys and paste your kie.ai API key
- 3
Start generating images and videos
Things to know
- Pre-built app is only available for macOS Apple Silicon; Windows and Linux need to be built from source.
- The macOS app is not code-signed, so Gatekeeper blocks it on first launch.
- Generation requires a paid kie.ai account and API key.
In collections
Topics
FAQ
- What is HeliosGen?
- A self-hosted desktop app for building AI image and video generation pipelines visually on an infinite node-based canvas, using your own kie.ai key. HeliosGen is a desktop app for chaining AI image and video generation steps into reusable workflows. You connect prompts, models, reference images and automation steps as nodes on an infinite canvas, and the pipeline can run in parallel or sequentially.
- Is HeliosGen free to use? What license does it have?
- HeliosGen has no declared license. Its code is public, but without a license you have no explicit permission to reuse it; ask the author before relying on it.
- Is HeliosGen still maintained?
- Updated within 6 months. The last commit to HeliosGen was on Sep 17, 2026.
- What can I use HeliosGen for?
- Producing a batch of images or videos from one prompt across several models Reusing a multi-step generation workflow with different reference images Keeping generated media and prompt history on a local machine
- How do I install HeliosGen?
- Download the macOS build from the Releases page Open Settings → API Keys and paste your kie.ai API key Start generating images and videos
- Where can I get HeliosGen?
- The source code and installation instructions are on GitHub: https://github.com/SegFault42/HeliosGen
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A web-based creative engine for generating and editing images with Stable Diffusion and other AI models, featuring a unified canvas and node-based workflows.
The official web front-end for ComfyUI, the node-based interface for running generative AI image and video workflows.
Open-source generative AI platform offering a unified API for text, image, video, audio, 3D and embeddings generation.
A collection of agent skills that teach AI coding agents to generate images, video, audio and 3D through the Scenario MCP server, plus skills that drive DCC tools and game engines.
Agent skills that let Claude Code, Cursor, Codex and similar coding agents generate images and video, train face identities, and build brand kits via Higgsfield AI.