OpenCut AI 0.1.8

OpenCut AI 0.1.8

中文官网 · English website · AI models, API keys and limits

• AI storyboards, ordered timeline placeholders, retries and optional last-frame continuity.
• Screen recording with separate system audio and compact pause/stop controls.
• Editable subtitles, track-wide font/color/shadow settings and image/text overlays.
• Preview/export effects, customizable transitions and visible timeline transition markers. Fixes subframe duration rounding at recording junctions.
• Sound downloads, local import and persistent timeline audio.

Downloads

• macOS Apple Silicon: Developer ID signed and Apple notarized DMG.
• Windows x64: unsigned NSIS installer. SmartScreen may display a warning; verify the release source and SHA256 checksum. Silent installation and packaged editor HTTP startup check passed on the GitHub Windows runner. Full editing/export and recording have not been verified on Windows.
• Linux x64: DEB and RPM built on GitHub Actions. Package metadata checks passed; the DEB's extracted application started successfully on Ubuntu 22.04. RPM installation, editing/export and screen recording have not been verified across distributions.

Use the attached OpenCutAI-SHA256SUMS.txt to verify files. Screen/system-audio recording and GPU behavior depend on the operating system and desktop environment.

AI currently uses ZenMux and requires a ZenMux API Key in Settings → ZenMux AI. Calls consume your provider account balance. MiniMax H3 Max requests support 5–15 seconds and 480p/768p in this application; other limits depend on the selected route/model. The website documents preset model IDs and their availability caveats. Prompts/reference media are sent to the provider; local editing needs no subscription.

Based on OpenCut Classic, with MIT license and original notices retained.