AMFORGE/GearCut

<div align="center">

<img src="https://files.manuscdn.com/useruploadbymodule/sessionfile/310519663479543323/yIhmCPspKLtzwUJf.png" alt="GearCut Interface" width="800"/>

✂️ GearCut

Natural Language Video Editor — Edit videos by describing what you want.

     

**Try the Demo →** · **AMFORGE Organization** · **Website** · **YouTube**

</div>

What is GearCut?

GearCut is a natural language video editing engine developed by AMFORGE. Instead of learning complex video editing software, you simply describe your edit in plain English — and GearCut's model translates your instruction into a structured list of editing operations that the project compiler then executes.

The core model (gc_editor) is built on AMFORGE's in-house SparseMind architecture — sparse attention, sparse FFN, dynamic neuron typing, and episodic memory. It contains 28,759,300 parameters (~28.8M) with a specialized vocabulary of 682 tokens designed exclusively for video editing semantics. It understands temporal references, clip identifiers, and export configurations, then generates a structured operation plan with frame-accurate precision.

"remove the first 3 seconds" → [{"op":"trim","clip":"c1","in":3.0,"out":8.0}] — done.

Key Features

Quick Start

Web Interface (Recommended)

The easiest way to use GearCut is through the live Gradio demo on this Space. Upload your video, type your instruction, and download the result.

<img src="https://files.manuscdn.com/useruploadbymodule/sessionfile/310519663479543323/NbmbzSLHdbkATMKE.png" alt="GearCut Web Interface Demo" width="750"/>

Local Installation

Requirements: Python 3.10+, ffmpeg installed and accessible in PATH.

# Clone the repository
git clone https://github.com/Volgat/gearcut
cd gearcut

# Install dependencies
pip install -r requirements.txt

# Launch the web interface
python gearcut_app.py --ui

# Or use the CLI directly
python gearcut_app.py "your instruction here" video.mp4

Project structure after setup:

<img src="https://files.manuscdn.com/useruploadbymodule/sessionfile/310519663479543323/UXPTqakKspKwLpSJ.png" alt="GearCut Project Structure" width="650"/>

CLI Usage

The command-line interface follows a simple pattern:

python gearcut_app.py "<instruction>" <video1.mp4> [video2.mp4 ...]

GearCut loads the model, parses your instruction, builds a timeline, and renders the output — all in a single command.

Example 1 — Remove the beginning

python gearcut_app.py "remove the first 3 seconds and export as out.mp4" test2.mp4

Output:

GearCut Editor v1-editor by AMFORGE: 9,721,219 params
gc_editor loaded (9,721,219 params, vocab=682)
timeline given to the model: clips: c1=test2.mp4(0.0-109.9)
operations: [{"op": "trim", "clip": "c1", "cut": "start", "seconds": 3.0},
             {"op": "export", "path": "output.mp4", "preset": "youtube_1080p"}]
timeline:
  c1: test2.mp4 [3.0 -> 109.9]  (106.9s)
rendering -> output.mp4
done: output.mp4  (106.9s)

<img src="https://files.manuscdn.com/useruploadbymodule/sessionfile/310519663479543323/tuOlkVtqKllpBvtS.png" alt="CLI Trim Demo" width="750"/>

Example 2 — Extract a specific range

python gearcut_app.py "keep only 120 to 600 of clip 1 and export as final.mp4" lab2.mp4

Output:

GearCut Editor v1-editor by AMFORGE: 9,721,219 params
gc_editor loaded (9,721,219 params, vocab=682)
timeline given to the model: clips: c1=lab2.mp4(0.0-848.5)
operations: [{"op": "trim", "clip": "c1", "cut": "range", "start": 120, "end": 600},
             {"op": "export", "path": "final.mp4", "preset": "youtube_1080p"}]
timeline:
  c1: lab2.mp4 [120.0 -> 600.0]  (480.0s)
rendering -> final.mp4
done: final.mp4  (480.0s)

<img src="https://files.manuscdn.com/useruploadbymodule/sessionfile/310519663479543323/hmVNppDceAPJHWSC.png" alt="CLI Range Demo" width="750"/>

Example 3 — Combine multiple clips

python gearcut_app.py "put final1.mov first then export" clip1.mp4 clip2.mp4

Output:

GearCut Editor v1-editor by AMFORGE: 9,721,219 params
gc_editor loaded (9,721,219 params, vocab=682)
operations: [{"op": "export", "path": "final1.mov", "preset": "youtube_1080p"}]
timeline:
  c1: clip1.mp4 [0.0 -> 15.015]  (15.015s)
  c2: clip2.mp4 [0.0 -> 15.015]  (15.015s)
rendering -> final1.mov
done: final1.mov  (30.033s)

<img src="https://files.manuscdn.com/useruploadbymodule/sessionfile/310519663479543323/kFgMBULZgmEoqFot.png" alt="CLI Combine Demo" width="750"/>

Web Interface Walkthrough

The Gradio interface provides the same capabilities as the CLI in a browser-friendly format. Upload one or more video files, enter your editing instruction in natural language, and click Submit. The processed video appears in the output panel for preview and download.

<img src="https://files.manuscdn.com/useruploadbymodule/sessionfile/310519663479543323/iXHxrnwfpRSOmCDB.png" alt="GearCut Gradio Editing" width="750"/>

The interface supports side-by-side comparison of the original and edited video, making it easy to verify that the edit matches your intent before downloading.

<img src="https://files.manuscdn.com/useruploadbymodule/sessionfile/310519663479543323/uAbLLWebiqYpeCxw.png" alt="GearCut Result Comparison" width="750"/>

Supported Instructions

GearCut understands a wide range of natural language patterns. The model was trained on video editing semantics and handles variations in phrasing naturally.

The grounding module validates parsed operations against the actual clip durations and corrects common misreads (e.g., filename mismatches, out-of-range timestamps) before rendering.

Architecture

GearCut is built around a lightweight transformer encoder-decoder architecture trained end-to-end on video editing instruction pairs.

Input: natural language instruction + clip metadata
         ↓
   [Tokenizer — vocab 682]
         ↓
   [gc_editor transformer — 9.7M params]
         ↓
   [Grounding module — validates against clip durations]
         ↓
   Structured operation plan (JSON)
         ↓
   [ffmpeg renderer — frame-accurate output]
         ↓
Output: edited video file

The tokenizer uses a custom vocabulary (gearcut_tok.vocab) optimized for temporal expressions, clip references, and export directives. The grounding module acts as a post-processing safety layer that rejects or corrects operations that would produce invalid results (e.g., trim beyond clip duration).

<img src="https://files.manuscdn.com/useruploadbymodule/sessionfile/310519663479543323/rcFAvaGqNPpySyHF.png" alt="GearCut Architecture Overview" width="750"/>

Model Details

The core model files (gearcut_compiler.py, gearcut_model.py, gearcut_infer.py, gearcut_ground.py) are proprietary and maintained in a private repository. The public interface (gearcut_app.py, gearcut_ui.py) downloads the compiled modules at runtime.

Evaluation

Measured on a held-out synthetic validation split. The meaningful metrics are not perplexity but whether the generated operations are directly usable:

Requirements

gradio>=5.0
torch>=2.0
sentencepiece
ffmpeg-python

ffmpeg must be installed separately and available in your system PATH. On Windows, place the ffmpeg binary in the gearcut/ffmpeg/ directory.

License

GearCut is released under the CKL License. See ameforge.tech for full terms. The model weights and core inference modules are proprietary assets of AMFORGE.

About AMFORGE

AMFORGE is an independent AI research studio focused on building efficient, practical AI systems. GearCut is part of a broader research direction exploring natural language interfaces for creative tools.

• —Website: ameforge.tech
• —YouTube: youtube.com/@ameforge1
• —GitHub: github.com/Volgat
• —Contact: contact@ameforge.tech

<div align="center"> _{Built with ❤️ by AMFORGE · <a href="https://ameforge.tech">ameforge.tech</a>} </div>