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

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.

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.

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.

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.

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.

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.

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.

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

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

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).

• FAL API Key: Get it from fal.ai/dashboard/keys
• Runware API Key: Sign up at runware.ai and get your key from the dashboard

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

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

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
  ],
}

The project includes safe update scripts that preserve existing configurations:

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

• fal.ai - AI model infrastructure
• Runware.ai - High-performance AI inference platform
• Next.js - React framework
• Remotion - Video processing and composition
• FFmpeg.wasm - Client-side video processing
• IndexedDB - Browser-based storage
• next-intl - Internationalization
• Vercel - Deployment platform

1. Clone the repository:

git clone https://github.com/timoncool/videosos
cd videosos

2. Install dependencies:

npm install

3. Start the development server:

npm run dev

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

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

2. Start services with Docker Compose:

docker compose up -d

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

To stop services:

docker compose down

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.

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.

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