Sora 2 CLI - OpenAI Sora 2 Video Generator Command Line Tool

A powerful command-line tool for generating and editing videos using OpenAI's Sora 2 and Sora 2 Pro API. Create stunning AI-generated videos from text prompts with Sora 2, the latest video generation model from OpenAI.

What is Sora 2?

Sora 2 is OpenAI's groundbreaking AI video generation model that creates high-quality videos from text descriptions. Sora 2 builds upon the original Sora model with improved video quality, better prompt understanding, and more realistic motion. With Sora 2 Pro, you can generate production-quality cinematic videos perfect for marketing, content creation, and creative projects.

This CLI tool provides easy access to Sora 2's powerful video generation capabilities directly from your terminal.

Features

• 🎬 Create Videos: Generate videos from text prompts using Sora 2 or Sora 2 Pro
• ✨ AI-Enhanced Prompts: Automatically improve prompts with GPT-5 for better results
• 🖼️ Image References: Use reference images as the first frame of your videos
• 🎨 Remix Videos: Edit existing videos with targeted adjustments
• 📊 Progress Tracking: Real-time progress bars during video generation
• 💾 Auto-Download: Automatically saves videos, thumbnails, and spritesheets

Installation

Prerequisites

• Python 3.11 or higher
• UV package manager
• OpenAI API key with Sora access

Setup

1. Clone or navigate to the project directory

2. Install dependencies with UV

   uv sync

3. Install the CLI tool

   uv pip install -e .

4. Configure your API key

   cp .env.example .env
   # Edit .env and add your OpenAI API key

   Get your API key from OpenAI Platform

Usage

Create Command

Generate new videos from text prompts.

sora-cli create --prompt "Your prompt here" --output-id video_name

Options

• --model / -m: Choose model (sora-2 or sora-2-pro, default: sora-2)
• --prompt / -p: Video generation prompt (required)
• --seconds / -s: Video duration in seconds (default: 5)
• --image / -i: Path to reference image (optional)
• --output-id / -o: Output filename without extension (required)
• --no-enhance: Skip GPT-5 prompt enhancement

Examples

Basic video generation:

sora-cli create -p "A cat riding a motorcycle through neon-lit streets" -o cat_motorcycle

Use Sora 2 Pro for higher quality:

sora-cli create --model sora-2-pro -p "Cinematic sunset over ocean waves" -s 10 -o sunset

With reference image:

sora-cli create -p "She turns around and smiles, then walks out of frame" -i reference.jpg -o walking

Skip prompt enhancement (use prompt as-is):

sora-cli create -p "Wide shot of desert highway, heat ripples, hard sun" -o desert --no-enhance

Edit Command

Remix existing videos with targeted modifications.

sora-cli edit --video-id video_abc123 --prompt "Change instructions" --output-id new_version

Options

• --video-id / -v: ID of the video to remix (required)
• --prompt / -p: Remix instructions (required)
• --output-id / -o: Output filename without extension (required)

Examples

Change color palette:

sora-cli edit -v video_abc123 -p "Shift to warm tones with golden hour lighting" -o warm_version

Modify subject:

sora-cli edit -v video_abc123 -p "Change the cat to orange color" -o orange_cat

Models

Sora 2 (sora-2)

• Best for: Speed and iteration
• Use cases: Prototyping, social media, quick feedback
• Characteristics: Fast generation, good quality

Sora 2 Pro (sora-2-pro)

• Best for: Production quality
• Use cases: Marketing, cinematic footage, high-fidelity content
• Characteristics: Longer generation time, higher cost, superior quality

Output Files

Videos are saved in the videos/ directory:

• {output-id}.mp4 - The generated video
• {output-id}_thumbnail.webp - Thumbnail image
• {output-id}_spritesheet.jpg - Spritesheet for scrubbing

Prompt Enhancement

By default, the CLI uses GPT-5 (o3-mini) to enhance your prompts based on Sora 2 best practices:

• Adds specific camera angles and movements
• Describes lighting and color details
• Clarifies subject actions and timing
• Improves overall cinematographic specificity

To skip enhancement and use your exact prompt, add the --no-enhance flag.

Tips for Better Results

1. Be specific: Describe shot type, subject, action, setting, and lighting
2. One action: Focus on one clear camera move and one subject action
3. Visual language: Use concrete details instead of vague descriptions
4. Lighting: Specify light sources and color tones
5. Remix carefully: Make single, well-defined changes for best results

Error Handling

• The tool validates API keys before running
• Progress bars show generation status
• Failed generations display error messages
• Missing files and invalid formats are caught early

Development

Project Structure

sora-video-cli/
├── .env                    # API key (not in git)
├── .env.example           # Template for .env
├── pyproject.toml         # UV project configuration
├── src/
│   └── sora_cli/
│       ├── cli.py         # Main CLI interface
│       ├── create.py      # Video creation logic
│       ├── edit.py        # Video remix logic
│       └── enhance.py     # Prompt enhancement
└── videos/                # Generated videos

Running in Development

# Activate virtual environment
source .venv/bin/activate

# Run CLI directly
python -m sora_cli.cli create -p "Test prompt" -o test

Resources

• OpenAI Sora API Documentation
• Sora 2 Prompting Guide
• Get API Key

License

MIT

---

Keywords

Sora 2, Sora 2 Pro, OpenAI Sora, Sora video generator, AI video generation, text to video, Sora API, Sora CLI, video generation CLI, OpenAI video AI, Sora 2 API, AI video creation, automated video generation, Sora command line tool