LibreAI / Video / printfilm
// video · yi1108/printfilm

printfilm

printfilm ist ein Open-Source-Projekt aus der Kategorie Video von yi1108 (yi1108/printfilm auf GitHub). Es hat 4.711 Sterne auf GitHub und 569 Forks, wird unter der Lizenz MIT veröffentlicht und wurde zuletzt am 8. Oktober 2026 auf GitHub aktualisiert.

PRINTFILM turns scripts and topics into finished AI short videos and animated short dramas

★ 4,7 Tsd. Sterne⑂ 569 ForksMITaktualisiert Okt 2026
▦ Visual — das README dieses Projekts enthält keinen einbettbaren Screenshot,
darum hier ein Platzhalter. Die Anleitung unten ist trotzdem vollständig.
// projektdaten

Projektdaten

GitHub-Sterne
4.711
Forks
569
Lizenz
MIT
Zuletzt auf GitHub aktualisiert
8. Oktober 2026
Kategorie
Video
Eintragstyp
Vollständige Installationsanleitung
Offizielle Website
Official site
// so funktioniert es

So funktioniert es

  1. Decide your content

    Choose the short-video pipeline (marketing/science clips from templates) or the drama pipeline (episodes from an outline or script).

  2. Clone and configure

    Clone the repo and copy the Docker environment example; set the database password, secret key, and your TokenFree API key (or mock mode to preview the UI).

  3. Start the stack

    Bring the full stack up with Docker Compose and wait for the API health check.

  4. Create your first project

    Register an account, pick a template or enter a script, generate storyboards, and export the finished film.

// funktionen

Warum es so beliebt ist

AI short-drama pipeline

From one line or an outline to episode scripts, parsed storyboards, and finished episodes, with a reusable asset library for characters, scenes, and props.

AI short-video pipeline

20+ built-in style templates; full (image + video + compose) or faster image_text modes, with per-shot redraw and regeneration.

Tool center

Standalone tools that skip the pipeline: text-to-image, image-to-image, product images, text-to-video, video-to-video, and e-commerce collages.

In-process task platform

A built-in scheduler/executor/poller runs generation jobs without a separate Celery worker, and jobs keep running when you leave the page.

Admin backend and open API

Manage users, orders, templates, tasks, and model routing; an open /api/v1 offers image and video generation with Bearer or X-Api-Key auth.

Configurable models and billing

Switch models from the admin backend without code changes; optional usage-based billing is off by default.

Advertisement
// selbst ausführen

Führe es selbst aus

  1. Clone the repository

    Clone the project and enter its directory.

    git clone https://github.com/yi1108/printfilm.git
    cd printfilm
  2. Prepare the environment

    Copy the Docker env example and edit it — set POSTGRES_PASSWORD, SECRET_KEY, and your TokenFree key (or ARK_MOCK=true to preview without a key).

    cp deploy/.env.docker.example deploy/.env.docker
  3. Start with Docker Compose

    Launch the full stack; the user app is on port 8080, admin on 8081, and the API on 8000.

    docker compose --env-file deploy/.env.docker up -d
// anwendungsfälle

Was du damit bauen kannst

Advertisement
// häufige fragen

Häufige Fragen

Who maintains PRINTFILM?

Developer yi1108; the hosted service runs at printfilm.com and the code is developed in the open on GitHub.

Is PRINTFILM free?

The code is MIT-licensed and free to self-host. Real generation needs a TokenFree API key, which is a paid third-party service; without a key you can only run the mock mode to view the interface.

Does it need a paid API key?

Yes for actual generation — text, image, and video all go through the TokenFree New API with one key (set as both OPENAI_API_KEY and ARK_API_KEY).

How does it compare to generic video generators?

Instead of one-off clips, PRINTFILM is an end-to-end production platform — script parsing, storyboards, reusable assets, batch task management, an admin backend, and an open API in one self-hosted stack.

How do I get started?

Install Docker, clone the repo, copy and edit deploy/.env.docker, run `docker compose --env-file deploy/.env.docker up -d`, then open http://localhost:8080 and register.

// weiter entdecken

Ähnliche Tools

Video

AIComicBuilder

Turn scripts into animated comics — AI characters, storyboards and video.

★ 1,9 Tsd. SterneApache-2.0
Anleitung ansehen →
Video

openshorts

Local-first pipeline that turns a topic into a publish-ready short video.

★ 32 SterneMIT
Anleitung ansehen →
Video

video-shotcraft

An agent skill that turns Claude Code or Codex into a motion-design studio for cinematic product videos.

★ 10,9 Tsd. SterneApache-2.0
Anleitung ansehen →
Video

synctoon

Python-based automated 2D animation utility that produces videos from text scripts

★ 50 SterneGPL-3.0
Anleitung ansehen →
Video

Jellyfish

An end-to-end production workspace for AI-generated short dramas. From script input to structured storyboarding, consistency management,…

★ 6,6 Tsd. SterneApache-2.0
Anleitung ansehen →
Video

Open-AI-Micro-Drama-Generator

Turn any idea into a short-form AI drama with a multi-agent pipeline.

★ 522 SterneCustom
Anleitung ansehen →
// alternativen

Alternativen im Bereich Video

Die anderen Projekte der Kategorie Video mit den meisten Sternen im LibreAI-Katalog — die realistische Auswahl, falls printfilm nicht das Richtige ist.

// mehr in video

Mehr in Video

Verwandte Projekte der Kategorie Video im Katalog, die printfilm bei den Sternen am nächsten stehen — zum Weiterstöbern auf beiden Seiten.

Hol dir printfilm

Kostenlos und offen — hol dir den Quellcode, gib dem Repository einen Stern und leg los.