Magiviz open-sources an end-to-end AI video creation suite
ItusiAI just open-sourced Magiviz, a turnkey platform that automates video production across five steps, from scriptwriting and character design to storyboarding, scene generation, and final stitching. It connects models like Veo, Kling, and Wan, and comes preloaded with Stripe billing and user management ready for deployment.
Magiviz is an open-source, end-to-end AI video studio that packs a full production pipeline into a single deployable web application. It automates every step from scriptwriting and character design to storyboarding, scene video rendering, and final assembly.
What it does
The engine runs on a five-step serial pipeline with state tracking and pause-and-resume support built into its core workflow.
Step one turns a user prompt, desired duration, and aspect ratio into a structured script with scene breakdowns, character profiles, and scene prompts.
Step two generates character visual sheets in parallel, supporting user-uploaded reference images for visual consistency.
Step three creates storyboard panels for each scene. It only passes relevant character art into the scene prompt and supports first-and-last frame generation.
Step four renders individual scene clips in parallel, supporting models like Veo 3.1, Kling 3.0, Seedance 2.0 and 2.5, Wan 3.0, and MiniMax H3.
Step five stitches all rendered scene clips and audio tracks into a cohesive final export using FAL AI.
The system supports dual-layer versioning. If one shot or character looks off, you can regenerate that specific element without losing past project history.
Beyond generation, Magiviz includes a ready-to-run SaaS infrastructure: NextAuth authentication, Stripe subscriptions and credit packs, referral links, and an admin console.
How to run it
The stack is built on Next.js 15, TypeScript, Tailwind CSS, Drizzle ORM, and PostgreSQL.
Because all generation requests route through external APIs like Kie.ai, ZenMux, and FAL, you do not need local GPUs to run the platform.
You only need Node.js 18+ and a working PostgreSQL database.
Clone the ItusiAI/Open-Magiviz repository from GitHub and run npm install.
Copy .env.example to .env.local, then supply your database URL, NextAuth secrets, Stripe credentials, and API keys for the AI providers.
Run npm run db:push to apply the schema, followed by npm run dev, or deploy directly to Vercel.
Why it matters
For indie hackers and lean teams looking to launch an AI video generation product, Magiviz eliminates weeks of plumbing work. You do not have to write multi-step polling state machines or stitch payment webhooks from scratch.
The main trade-off is that it relies heavily on third-party API aggregators out of the box. If you want to wire it to your own ComfyUI instances or private GPUs, you will need to rebuild parts of the backend routing. But as a monetization-ready launchpad, it is remarkably complete.
Sources (1)
- github.com(opens in a new tab)Originalgithub.com