Clips-Studio / llms.txt
clips-studio-bot
Sync website from f4bbeac
c89844a
Raw History Blame Contribute Delete
8.96 kB
# Clips Kitty
> Free, open-source AI video clipping software and Opus Clip alternative that
> runs entirely on your own Windows PC.
> Paste a Twitch VOD, Kick VOD or YouTube link and it finds the moments worth
> posting, crops them to vertical with the speaker kept in frame, burns in
> word-synced captions and writes the titles. No uploads, no subscription, no
> per-clip fee. AGPL-3.0.
Clips Kitty is a desktop application, not a web service. Everything (the
transcription, the clip selection, the subject tracking and the language model)
runs on the user's own machine by default. When the PC is too old, OpenRouter,
signed in with the user's own account, can run the language model or the
transcription instead. It is a local-first, open-source alternative to
Opus Clip and other cloud clipping tools, aimed at creators who would rather not
pay per video or upload their footage to someone else's server.
## Mission
Clips Kitty is a free, open-source AI video editor built to help small
streamers and creators turn their long-form videos and livestreams into
short-form content without hiring an editor or spending hours editing footage
themselves.
## How it compares to Opus Clip
- **Price:** free, AGPL-3.0, no paid tier. Opus Clip has a free tier plus
$15/mo and $29/mo plans (checked 12 September 2026).
- **Uploads:** none. The video is read where it sits. Opus Clip uploads the
whole video to its servers before processing.
- **Limits:** no clip cap, no watermark, no export expiry. Opus Clip's free
tier watermarks, caps at 1080p and expires exports after three days.
- **The catch:** Clips Kitty needs Windows, 16 GB of RAM and ideally an NVIDIA
GPU. Opus Clip runs in any browser, including on a phone. If you have no
suitable PC, or you clip only occasionally, Opus Clip is the better choice.
## What it actually does
- **Transcribes** with faster-whisper at word level, so captions land on the
right syllable and clips start on real sentence boundaries.
- **Scores** candidate moments with a local language model reading the
transcript alongside audio and visual signals: loudness, laughter-shaped
bursts, scene cuts, motion, on-screen reactions.
- **Reframes** to 9:16 with YOLO pose tracking and active-speaker detection, so
the person talking stays in shot and it cuts between speakers rather than
zooming out to hold both.
- **Frames matches and game streams**: a soccer match follows the ball (the
Sports toggle); Gaming / Reaction puts the webcam and the game together in one
of eleven layouts; a stream that was 9:16 all along keeps its own layout
(Vertical Live).
- **Watches channels**: new videos on a YouTube, Twitch or Kick channel are
spotted (YouTube's through its RSS feed) and clipped by themselves.
- **Takes requests in a sentence**: the Ask Clips Kitty box at the bottom of the
app queues a video, takes what the clips should be about and how they should
look. It needs a Gemma 4 model (`gemma4:e4b`, or `gemma4:e2b` under 6 GB of
VRAM), installed from the app's Models page, or the cloud model the user chose.
- **Posts, if asked**: clips go to nine platforms at once through a publishing
service the user connects, now, on a schedule, or hands-off for a watched
channel.
- **Edits**: a built-in editor for fixing anything the AI got wrong, caption
styling, watermarks, multilingual captions and dubbing.
## What it is good and bad at
- **Good at:** IRL, just chatting, podcasts, vlogs, interviews. That is what it
is tuned for and tested on.
- **Also does:** game streams and reaction videos (Gaming / Reaction), and
soccer matches (Sports), with goals read from the score box, the crowd and the
commentary.
- **Not yet:** basketball and cricket, next on the roadmap. VTubers are not
supported: the framing needs a real person on camera.
## Requirements
Windows 10 or 11, 64-bit. 16 GB RAM recommended, ~20 GB free disk. An NVIDIA
GPU is strongly recommended. It runs on CPU, just far slower. One installer
carries the app, the engine, FFmpeg, the Ollama runtime and the tracking and
transcription models; the language model downloads on first launch, sized to
the graphics card.
Install it from the Microsoft Store, with the Web Setup from the GitHub releases,
or from a terminal: `winget install ColinGPT9.ClipsKitty`.
## For developers
The Clips Kitty SDK (`clipskitty_sdk`, Python, MIT) is for writing a small
plugin that finds, understands or rates the moments in a video, for one game
or one kind of show; Clips Kitty does the rest: the download, the transcript,
and cutting, framing and captioning the clips. Every step is optional: one
plugin can find the moments, and up to 3 can understand and up to 3 can rate
them. Up to 3 can suggest edits for the clips, which wait for the creator in
the editor. Posting isn't a plugin step: Clips Kitty posts clips itself, and
through WoopSocial (https://woopsocial.com/?via=clipskitty) it can post to
many sites at once on the creator's own account. Affiliate link - Clips Kitty
may earn a commission if you sign up through it, at no extra cost to you.
A plugin runs on Clips Kitty's own Python 3.11, with the standard
library (minus a few modules it leaves out) and the SDK, on the creator's PC
with the creator's rights, like any program: Clips Kitty doesn't sandbox it,
and its install screen shows what the plugin declares. The templates and
tutorial use Quarkbloom Arena, a made-up game.
No Clips Kitty release runs plugins yet. Plugins made with SDK 1.2.0 need the
first release that includes it; until that is out, run Clips Kitty from source.
- [Clips Kitty SDK](https://colingpt9.github.io/clips-studio/developers.html)
- [Your first game pipeline](https://github.com/ColinGPT9/clips-studio/blob/main/docs/developers/first-game-pipeline.md)
- [The SDK's README](https://github.com/ColinGPT9/clips-studio/blob/main/sdk/python/README.md)
- [Developer docs](https://github.com/ColinGPT9/clips-studio/blob/main/docs/developers/README.md)
## Links
- [Website](https://colingpt9.github.io/clips-studio/)
- [Source on GitHub](https://github.com/ColinGPT9/clips-studio)
- [Download](https://github.com/ColinGPT9/clips-studio/releases)
- [Clips Kitty Web](https://clips-kitty-web.vercel.app/): the clip picking in a
browser, on a Twitch or Kick VOD or a local file, billed to the visitor's own
OpenRouter account. No vertical framing or captions; those are the desktop app.
- [Architecture](https://github.com/ColinGPT9/clips-studio/blob/main/ARCHITECTURE.md)
- [Changelog](https://colingpt9.github.io/clips-studio/changelog.html)
- [Roadmap](https://colingpt9.github.io/clips-studio/roadmap.html)
- [Known issues](https://github.com/ColinGPT9/clips-studio/blob/main/KNOWN-ISSUES.md)
- [Twitch clip generator: Twitch VODs into TikTok clips, Shorts and Reels](https://colingpt9.github.io/clips-studio/twitch.html)
- [Kick clip generator: Kick VODs into TikTok clips, Shorts and Reels](https://colingpt9.github.io/clips-studio/kick.html)
- [Clipping YouTube](https://colingpt9.github.io/clips-studio/youtube.html)
- [Game stream clips](https://colingpt9.github.io/clips-studio/gaming.html)
- [Soccer highlight clips](https://colingpt9.github.io/clips-studio/sports.html)
- [Local vs cloud clipping](https://colingpt9.github.io/clips-studio/local-vs-cloud.html)
- [Opus Clip alternative](https://colingpt9.github.io/clips-studio/opus-clip-alternative.html)
- [Descript alternative](https://colingpt9.github.io/clips-studio/descript-alternative.html)
## In other languages
The same overview, written in each language the app supports:
[Español](https://colingpt9.github.io/clips-studio/es.html) ·
[Português](https://colingpt9.github.io/clips-studio/pt.html) ·
[Français](https://colingpt9.github.io/clips-studio/fr.html) ·
[Deutsch](https://colingpt9.github.io/clips-studio/de.html) ·
[हिन्दी](https://colingpt9.github.io/clips-studio/hi.html) ·
[Bahasa Indonesia](https://colingpt9.github.io/clips-studio/id.html) ·
[日本語](https://colingpt9.github.io/clips-studio/ja.html) ·
[Русский](https://colingpt9.github.io/clips-studio/ru.html) ·
[العربية](https://colingpt9.github.io/clips-studio/ar.html) ·
[简体中文](https://colingpt9.github.io/clips-studio/zh.html) ·
[Tiếng Việt](https://colingpt9.github.io/clips-studio/vi.html) ·
[Filipino](https://colingpt9.github.io/clips-studio/tl.html) ·
[Türkçe](https://colingpt9.github.io/clips-studio/tr.html) ·
[اردو](https://colingpt9.github.io/clips-studio/ur.html) ·
[বাংলা](https://colingpt9.github.io/clips-studio/bn.html) ·
[ไทย](https://colingpt9.github.io/clips-studio/th.html) ·
[한국어](https://colingpt9.github.io/clips-studio/ko.html) ·
[Italiano](https://colingpt9.github.io/clips-studio/it.html)
## Status
Alpha. It works and is used daily, but it is early and has rough edges, which
are listed honestly in the known-issues document rather than discovered by
users.