Edit videos by telling it what you want.

One sentence → scene detection, smart cuts, auto subtitles, color grading. Plus: learn any video's editing style and reuse it.

$ pip install cutai
Download macOS App

No timeline dragging. Just ask.

Describe your edits in plain English or Korean. CutAI analyzes your video, detects scene boundaries, and translates your intent into a concrete edit plan.

  • Scene detection & silence trimming
  • Auto subtitles via Whisper
  • Color grading (warm, cinematic, cool)
  • Background music mixing

$ cutai plan vlog.mp4 -i "cut dead air and add subtitles"

🎬 Analyzing vlog.mp4...
✓ Detected 14 scenes
✓ Transcribed speech
✓ Found 3 low-value segments
📋 Edit Plan:
• Remove dead air (0:12-0:15, 0:45-0:48)
• Add burned-in subtitles
• Apply neutral color grade
Edit Style Transfer (EST)

Steal any YouTuber's editing style.

Extract the editing patterns from a reference video and apply them to your own footage. It's like style transfer, but for editing decisions.

1

Pick a reference

Give CutAI a video from a creator whose style you admire.

2

Extract Edit DNA

It automatically analyzes pacing, cuts, colors, and subtitle patterns.

3

Apply to your video

Render your raw footage using the extracted style.

# Extract style from MKBHD

$ cutai style-extract reference.mp4 -o style.yaml

rhythm:
  mean_scene_duration: 2.4
  pacing_curve: "dynamic"
visual:
  temperature: "warm"
subtitles:
  presence_ratio: 0.85

What does Edit DNA capture?

  • ✂️ Cut Rhythm Scene durations and pacing curves
  • 🎨 Visual Style Color temperature, brightness, saturation
  • 📝 Subtitles Timing patterns and presence
  • 🎵 Audio Mixing BGM presence and volume balance

Prefer a GUI? Try the Desktop App.

A native macOS app built with Tauri. Drag and drop videos, pick a style preset, and render locally. No command line required.

Download for macOS (Alpha)

Requires macOS 12+ (Apple Silicon)