AVPrep3 for Windows

Or the portable zip (339 MB, runs without installing).
SHA-256 8219a588cb2fb9d50b8b74833835a7463242e426cca0f7f7dc0ba48cdb9f3e06

Both downloads contain everything AVPrep3 needs and a README.txt with these installation steps.

Install script: double-click Install-AVPrep3.cmd to download AVPrep3, unzip it to your account, and add Start menu and desktop shortcuts. It needs no administrator rights and removes nothing. Add -Setup in PowerShell to run the Setup program instead.

Install

1. Download AVPrep3-Setup-0.1.0.exe and double-click it.
2. Windows SmartScreen will probably say it protected your PC, because the installer isn't code-signed. Click More info, then Run anyway.
3. Click through the installer. It installs for your Windows account only, in %LOCALAPPDATA%\Programs\AVPrep3, and needs no administrator rights.
4. Start it from the Start menu or the desktop shortcut.

Updating: close AVPrep3, then run the newer installer over the old copy. Your settings, presets and autosaves stay where they are.

Removing: Windows Settings, Apps, AVPrep3, Uninstall (or the Uninstall shortcut in its Start menu folder). The program goes; your settings, presets and autosaves are left in the data folder so nothing you made disappears by accident.

Portable copy instead

The zip runs without installing, from any folder you can write to, including a USB stick. Right-click the zip, choose Extract All, and extract the whole zip (don't run it from inside the zip viewer). Then double-click AVPrep3.exe in the extracted folder. If SmartScreen warns, click More info, then Run anyway. Keep the folder together; settings live in the data folder beside the program and travel with it. If Windows blocks the extracted files, open PowerShell in that folder and run Get-ChildItem -Recurse | Unblock-File.

What it needs

• Windows 10 or 11, 64-bit.
• About 800 MB of disk space, plus room for your exports.
• Nothing else. Everything is included: FFmpeg and FFprobe for import and export, the Whisper speech engine and its English model for captions, and the Microsoft Visual C++ runtime files Whisper needs. You don't need to install Node.js, Python, FFmpeg or any runtime. No account, no internet connection, nothing is uploaded.

What it does

• Unlimited video and audio tracks. Drag to move, drag the edges to trim, razor and split, undo to 100 steps.
• Real-time preview, crossfades, colour and transform per clip, text, images and shapes as clips.
• Audio mixing with EQ and ducking, so music drops under the voice on its own.
• Captions transcribed on your computer with Whisper, then edited and burned in or exported.
• Twelve platform presets (Reels, Shorts, TikTok, Stories and more), export to a target file size, metadata stripping, watermark, GIF, batch and variants.
• Fit to Length, images to video, silence auto-cut, loop finder, waveform visualizer, freeze frame and split export.

Quick start

1. Open Media, or drag video, audio and images into the window.
2. Drag them onto the timeline. Video lanes stack bottom to top; audio lanes mix.
3. Add Title for a three-second title, then place it in Preview.
4. Pick a Platform preset such as TikTok 9:16 for a vertical video. The canvas switches to 1080 × 1920 to match and titles are fitted inside; Undo puts it back.
5. Export. Pick a folder and a name, then Export Video.

Space plays and pauses. S splits at the playhead. Delete removes the selection. Ctrl+Z undoes, Ctrl+S saves the project, Ctrl+E opens Export. Help has the full keyboard map. Exports are always new files; your source media is never overwritten.