Skip to content

timoncool/videosos

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

VideoSOS - Open-Source AI Video Editor with 100+ Models

GitHub stars GitHub forks GitHub issues License: MIT Next.js Remotion

Create professional videos with AI in your browser. Free, open-source video editor featuring Google Veo 3.1, FLUX, Gemini 2.5 Flash, Imagen 4, and 100+ top AI models for text-to-video, image-to-video, text-to-image generation, image editing, music composition, and voiceover creation. No uploads, complete privacy, advanced timeline editing.

🚀 Try the Live Demo | 🇷🇺 Русская версия README

Screenshots

Video Editor — Main Interface

VideoSOS Video Editor - Main Interface

Video Editor — Project Statistics

VideoSOS Video Editor - Project Statistics


Key Features

🤖 100+ AI Models

VideoSOS integrates with fal.ai and Runware.ai to bring you cutting-edge AI generation capabilities:

  • 🎬 Video Generation: Text-to-video and image-to-video with state-of-the-art models including Google Veo 3.1, Kling 2.5, and Hailuo-02
  • 🎨 Image Generation: Text-to-image with cutting-edge models including Gemini 2.5 Flash, Imagen 4.0, FLUX.1 Pro, and Ideogram V3
  • ✏️ Image Editing: AI-powered photo editing with leading models including Gemini 2.5 Flash Edit, FLUX.1 Kontext, and Qwen Image Edit
  • 🎵 Audio Generation: Music composition with Stable Audio and natural voiceovers with multiple voices

Switch between providers seamlessly or use both simultaneously for maximum flexibility. See complete model list below.

🔒 100% Privacy-First Architecture

Everything runs locally in your browser. Your projects, media, and API keys are stored in IndexedDB on your device. No uploads, no tracking, no data collection.

🎬 Timeline Editor

Advanced multi-track video editing with drag-and-drop functionality, precise keyframe control, audio waveform visualization, and real-time preview. Support for multiple aspect ratios (16:9, 9:16, 1:1) perfect for social media platforms.

💰 Cost Tracking & Project Statistics

Built-in cost tracking and comprehensive project analytics help you stay within budget. Track generation costs per media item, view total project expenses, and get detailed breakdowns by media type, AI provider, and model usage. Perfect for managing your AI generation budget and analyzing usage patterns.

🔍 Advanced Model Selection

Enhanced model selection interface with intelligent search, filtering by provider (FAL/Runware), and categorization by type (text-to-video, image-to-video, text-to-image, image editing). Real-time pricing display for FAL models helps you make informed decisions before generating content.

🎨 Multi-Modal Content Generation

Create complete video projects with AI-generated images, videos, background music, and voiceovers. Advanced features include camera movement controls, video upscaling with Topaz, and automatic lip-sync for voiceovers.

⚡ Client-Side Video Processing

Powered by FFmpeg.wasm and Remotion for high-quality video rendering entirely in the browser. No server uploads, no waiting in queues. Export videos with proper audio mixing and precise timing control.

🌍 International & Accessible

Full internationalization support (English/Russian) with keyboard shortcuts for power users. Intuitive UI designed for both beginners and professionals.

Supported AI Models

VideoSOS features 100+ AI models across video generation, image generation, and image editing. All models are available through dual provider support for maximum reliability.

Complete Model List by Category

Text-to-Video Models

Model Provider Description
Veo 3.1 FAL, Runware Google's latest video model with native audio
Veo 3.1 Fast FAL, Runware Faster version of Veo 3.1
Veo 3 FAL, Runware Cinematic video with synchronized audio
Veo 3 Fast FAL Fast cinematic video generation
Kling 2.5 FAL, Runware High-quality video synthesis
Hailuo-02 FAL, Runware Advanced video generation

Image-to-Video Models

Model Provider Description
Veo 3.1 Image-to-Video FAL Animate images with Veo 3.1
Veo 3.1 Fast Image-to-Video FAL Fast image animation
Veo 3 Image-to-Video FAL, Runware Animate static images
Veo 3 Fast Image-to-Video FAL, Runware Fast image animation

Text-to-Image Models

Model Provider Description
Gemini 2.5 Flash Image FAL, Runware Ultra-fast image generation
Imagen 4.0 Ultra FAL, Runware Google's highest quality image model
Imagen 4.0 Preview Fast FAL, Runware Fast high-quality images
Hunyuan Image 3.0 FAL, Runware State-of-the-art image synthesis
Seedream 4.0 FAL, Runware High-resolution image generation
Seedream 3.0 FAL, Runware Quality image synthesis
GPT Image 1 FAL, Runware OpenAI's image model
FLUX.1 Pro FAL, Runware Professional image generation
Ideogram V3 FAL, Runware Text rendering in images

Image Editing Models

Model Provider Description
Gemini 2.5 Flash Edit FAL, Runware High-quality image editing
FLUX.1 Kontext Pro FAL, Runware Context-aware editing
FLUX.1 Kontext Dev FAL, Runware Developer-friendly editing
FLUX.1 Kontext Max FAL, Runware Maximum quality editing
Qwen Image Edit FAL, Runware Intelligent image modification
Reve v1 Edit FAL Advanced editing capabilities
Seedream 4.0 Edit FAL, Runware High-quality transformations
GPT Image 1 Edit FAL, Runware OpenAI image editing

AI Providers

VideoSOS supports multiple AI providers simultaneously:

  • fal.ai - Comprehensive AI model infrastructure with 50+ models for image, video, and audio generation
  • Runware.ai - High-performance AI inference platform with 100+ models including FLUX, Google Imagen, Ideogram, Kling, and more

You can use either provider independently or both simultaneously. Simply add your API keys in the Settings dialog (click the gear icon in the app).

Getting API Keys

Both API keys are stored locally in your browser and never sent to our servers.

For Developers

Updating Model Information

VideoSOS includes automated scripts to keep model information up-to-date with provider APIs. Here's how to update models:

FAL Models

Run the FAL model update script to fetch latest models and pricing:

npm run update-fal-models

This script:

  • Fetches all available models from FAL API
  • Updates model schemas and parameters
  • Syncs pricing information
  • Updates src/lib/fal.ts and related files

Runware Models

For Runware models, manually enrich model configurations in src/lib/runware-models.ts based on Runware documentation:

  1. Check model documentation for available parameters
  2. Add availableDimensions, hasNegativePrompt, availableSteps, etc.
  3. Verify changes match actual API capabilities

Example enrichment:

{
  endpointId: "google:4@1",
  label: "Gemini Flash Image 2.5",
  hasNegativePrompt: true,
  availableDimensions: [
    { width: 1024, height: 1024, label: "1024×1024 (1:1)" },
    { width: 1824, height: 1024, label: "1824×1024 (16:9)" },
    // ... more dimensions
  ],
}

Safe Updates

The project includes safe update scripts that preserve existing configurations:

# Apply FAL parameter updates safely
npm run safe-apply-parameters

Tech Stack

Quick Start

  1. Clone the repository:
git clone https://github.com/timoncool/videosos
cd videosos
  1. Install dependencies:
npm install
  1. Start the development server:
npm run dev

Open http://localhost:3000 to see the application.

Running with Docker

If you have Docker installed, you can run the project in a container:

  1. Clone the repository:
git clone https://github.com/timoncool/videosos
cd videosos
  1. Start services with Docker Compose:
docker compose up -d
  1. Open http://localhost:3000 to see the application.

To stop services:

docker compose down

Star History

Star History Chart

Contributing

We welcome contributions from developers of all skill levels! Whether you're fixing bugs, adding features, improving documentation, or sharing ideas, your help makes VideoSOS better for everyone.

How to contribute:

  1. ⭐ Star this repository to show your support
  2. 🐛 Report bugs or suggest features in Issues
  3. 🔧 Submit pull requests - see our Contributing Guide
  4. 📖 Improve documentation
  5. 💬 Join discussions and help other users

Every contribution counts, no matter how small! Let's build the best open-source AI video editor together.

Credits

Assembled by Nerual Dreming - founder of ArtGeneration.me, tech blogger, and neuro-evangelist.

This project was built with Devin, the AI software engineer. Experience autonomous coding with Devin's free trial.

This project is a fork of the original AI Video Starter Kit from the fal.ai community.

License

This project is licensed under the MIT License - see the LICENSE file for details.

About

Enable AI models for video production in the browser

Resources

License

Code of conduct

Contributing

Stars

Watchers

Forks

Packages

No packages published

Languages

  • TypeScript 91.6%
  • Python 4.6%
  • Batchfile 1.9%
  • Shell 1.0%
  • CSS 0.5%
  • Makefile 0.3%
  • Other 0.1%