VideoHighlighter Aseiel/VideoHighlighter

Local desktop app that scores the strongest moments in long footage from motion, audio, objects and transcript, then exports a highlight reel and clips.

Creator kit + Compare GitHub ↗ Install guide (official) ↗

VideoHighlighter at a glance

Checked 15 Oct 2026

VideoHighlighter (v0.13.3, October 2026) is a local desktop app for Windows, macOS and Linux that scores moments in long video (VODs, CCTV, GoPro, sports, interviews) from scene, motion, audio, object, action and transcript signals, explains each score in a report, and exports a highlight reel plus individual clips. Runs on CPU or Intel, AMD and NVIDIA GPUs; no API key needed. Code is AGPL-3.0; a separate paid Pro edition adds live overlays. It does not reframe to 9:16, burn styled captions, give a virality score or post to social platforms.

Cite this pageRepoQuarry. “VideoHighlighter.” Checked 15 Oct 2026. https://repoquarry.dev/projects/videohighlighter

How much of Opus Clip does it replace?

1 of 5 jobs. Finds and exports highlight clips from long footage; no vertical reframe, styled captions, virality score or social posting.

AI picks highlight clips from a long video Partial replacement

Auto-reframe to vertical 9:16 (speaker or face tracking) Not covered

Animated burned-in captions on clips Not covered

Virality score per clip Not covered

Schedule and publish clips to social platforms Not covered

For auto-reframe to vertical 9:16 (speaker or face tracking), animated burned-in captions on clips, virality score per clip, schedule and publish clips to social platforms, see other Opus Clip alternatives. You may also need to build an interface, an API server and hosting yourself.

Facts, with sources

Every answer links to where we found it. We read the documentation; we did not install it.

Report a correction

QuestionAnswerSourceChecked

Runs on my machine? Yes, local desktop app; nothing is uploaded Nothing is uploaded for analysis, and the basic pipeline needs no API key. github.com · VideoHighlighter#readme ↗ 1d ago

Code licence AGPL-3.0 (copyleft); also offered under a separate commercial licence github.com · LICENSE ↗ 1d ago

Model weights licence Stock object detector is YOLOX (Apache-2.0) and transcription is OpenAI Whisper (MIT); licences of other bundled models not listed Object detection is YOLOX (Apache-2.0): the runtime is OpenVINO above, and the stock models download on first use. github.com · requirements.txt ↗ 1d ago

Commercial use Yes under AGPL-3.0 conditions; a separate commercial licence is available from the copyright holders VideoHighlighter is also offered under a separate commercial license by the copyright holders. github.com · VideoHighlighter#license ↗ 1d ago

Paid services needed? None required; a separate paid Pro edition and your own hosted-service keys are optional This project ships no paid API keys. Bring your own if you want to use a hosted service. github.com · INSTALL.md ↗ 1d ago

Hardware CPU works; GPU optional (tested on Intel Arc A750, GTX 1060, Radeon RX 570); RAM not stated; Windows install about 3 GB VideoHighlighter runs on Intel, AMD and NVIDIA GPUs, and on CPU alone if there is nothing else. github.com · INSTALL.md ↗ 1d ago

Operating systems Windows and macOS builds (Mac less tested); Linux from source Free and open source (AGPL-3.0). Windows, macOS, Linux. Intel, AMD and NVIDIA GPUs. github.com · VideoHighlighter#install ↗ 1d ago

Setup effort Easy on Windows and macOS (unsigned installer); Linux runs from source with pip Per-user, no admin. It is not code-signed yet github.com · VideoHighlighter#install ↗ 1d ago