Documentation

CogVideo by N3uralia Platform

Professional AI video generation powered by CogVideoX with N3uralia's cinematic enhancement system

Getting Started

Learn how to create your first video

Prompt Guide

Master the art of video prompts

API Reference

Technical documentation

Getting Started

1. Choose Your Model

Select from our available CogVideoX models. We recommend CogVideoX-5B for the best quality and detail.

  • CogVideoX-5B: Best quality, highest detail (Recommended)
  • CogVideoX-2B: Faster generation, good quality

2. Write Your Prompt

Describe the video you want to create. Be specific about the scene, action, lighting, and camera movement.

"A woman walks through a neon-lit Tokyo street at night, camera tracking alongside her"

3. Enhance with N3uralia Style

Click "Enhance Prompt" to transform your prompt with our professional N3uralia cinematic enhancement system. This adds:

  • Professional camera movements and angles
  • Cinematic lighting descriptions
  • Atmospheric details and mood
  • Technical specifications optimized for CogVideoX

4. Generate Your Video

Click "Generate Video" and wait for the AI to create your video. Generation typically takes 2-5 minutes depending on the model.

Prompt Engineering Guide

Best Practices

  • Be Descriptive: Use 60-100 words for optimal results with CogVideoX
  • Include Camera Movement: Specify tracking, panning, zooming, or static shots
  • Describe Lighting: Golden hour, soft light, dramatic shadows, neon glow
  • Set the Scene: Time of day, weather, location, atmosphere
  • Define Action: What's happening in the scene, movement, interactions

Example Prompts

Nature Scene

"Drone view of waves crashing against rugged cliffs at sunset. The golden light illuminates the rocky shore as white-tipped waves create dramatic splashes. Camera slowly pulls back to reveal the vast coastline."

Urban Scene

"A stylish woman walks down a Tokyo street filled with warm glowing neon signs. She wears a black leather jacket and red dress. Camera tracks alongside her as she walks confidently through the bustling night scene."

Cinematic Scene

"Close-up of a mechanic's hands repairing a vintage motorcycle engine in a dimly lit garage. Warm overhead lights cast dramatic shadows. Tools and oil-stained rags surround the workspace. Camera slowly circles the subject."

N3uralia Enhancement System

Our proprietary enhancement system transforms simple prompts into professional-grade descriptions by incorporating:

Technical Elements

• Camera angles and movements

• Focal length specifications

• Lighting setups

• Composition rules

Artistic Elements

• Atmospheric details

• Color grading notes

• Mood and emotion

• Cinematic references

API Reference

Enhance Prompt Endpoint

POST /api/enhance-prompt

Enhances a simple prompt with N3uralia's professional cinematic style.

{
  "prompt": "A woman walking in Tokyo"
}

Generate Video Endpoint

POST /api/generate

Generates a video using CogVideoX with the provided prompt.

{
  "prompt": "Enhanced prompt text...",
  "model": "cogvideox-5b",
  "title": "My Video",
  "description": "Description",
  "tags": ["tokyo", "cinematic"]
}

About CogVideoX

CogVideoX is a state-of-the-art text-to-video generation model developed by THUDM. It uses advanced diffusion models to create high-quality videos from text descriptions.

Key Features

• High-quality video generation up to 6 seconds

• Support for complex scenes and camera movements

• Excellent understanding of natural language prompts

• Consistent motion and temporal coherence

• Multiple model sizes for different use cases

About N3uralia

N3uralia is a creative technology group specializing in AI-powered media generation and enhancement. This platform represents our commitment to making professional-grade video generation accessible to everyone.

Our Mission

We believe that powerful creative tools should be accessible, intuitive, and produce professional results. Our enhancement system transforms simple ideas into cinematic masterpieces, bridging the gap between imagination and reality.