A free, open-source, modular, node-based tool designed to streamline AI-powered video production. It addresses the complexity of orchestrating multiple AI APIs for multi-shot video generation by providing a visual interface, automating workflow steps, and significantly boosting generation efficiency for creators.

How It Works

The core of AI Film Studio is a modular, node-based interface akin to professional video editing software like Maya or Nuke. Users construct video production pipelines by connecting nodes representing distinct tasks such as script input, storyboard generation, image/video description, keyframe generation, and final video rendering. This approach allows for independent editing of API parameters and cached content for each node. The system intelligently automates connections and execution flows between many nodes, reducing manual effort and integrating seamlessly with various third-party AI APIs (e.g., OpenAI, Gemini) and local ComfyUI instances for advanced image and video synthesis.

Quick Start & Requirements

1. Installation:
   • Ensure Python is installed and create a virtual environment (venv).
   • Navigate to the project directory in your terminal.
   • Run pip install -r requirements.txt.
2. Launch:
   • In the project terminal, run python app.py.
   • Access the web interface at http://127.0.0.1:5000.
3. Prerequisites:
   • Python 3.x.
   • Dependencies listed in requirements.txt.
   • API Keys and Addresses for desired LLM and video/image generation services (e.g., OpenAI, Gemini, Volcengine, Google AI Studio).
   • Optional: A locally running ComfyUI instance is required for the keyframe generation module.
4. Configuration: Within the web UI, configure LLM API Address, LLM API Key, LLM Model Name, and optionally the ComfyUI Address.
5. Resources: Refer to the Bilibili video guide for detailed configuration and usage: https://www.bilibili.com/video/BV1fxPWz4EPJ/.

Highlighted Details

• Modular Node Interface: Visual, drag-and-drop workflow creation for AI video pipelines.
• Automated Workflow: Intelligent auto-linking and execution between sequential nodes (e.g., script to storyboard to description).
• API Integration: Supports multiple LLM and generative AI APIs, with configurable parameters per node.
• Global Presets: Includes presets for image and video generation, integrating with Volcengine and Google AI Studio.
• Node Status Indicators: Visual cues (color-coded) for node execution status (pending, processing, completed, failed, approved).
• ComfyUI Integration: Dedicated module for keyframe generation requires a local ComfyUI setup with a provided workflow JSON.

Maintenance & Community

This project is community-maintained, actively welcoming code contributions, issue submissions, and discussions. Development principles emphasize transparency and openness.

Licensing & Compatibility

The project is released under the MIT License, permitting free use, modification, and distribution, provided the original copyright notice is retained. This license is generally compatible with commercial use and integration into closed-source projects.

Limitations & Caveats

Users are solely responsible for any costs incurred from third-party API calls and for the security of their data. The tool's primary benefit is enhancing generation efficiency and workflow automation, not necessarily improving the inherent quality of the AI-generated video content. Keyframe generation is dependent on a correctly configured local ComfyUI instance.