# AI use cases for Autoclip (artbyjazi/autoclip)

Repo: [artbyjazi/autoclip](https://github.com/artbyjazi/autoclip) | Use cases: 3 | Page: https://dedupedai.com/use-cases/artbyjazi-autoclip/

AutoClip is a local-first Python app that turns a long video into ranked, captioned, speaker-tracked 9:16 clips using Whisper plus Ollama or your own LLM key.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Turn a long video into captioned vertical clips offline: https://dedupedai.com/use-cases/artbyjazi-autoclip/clip-long-video-into-vertical-shorts/
- Use your own LLM key for sharper clip selection: https://dedupedai.com/use-cases/artbyjazi-autoclip/better-clip-picks-with-your-own-key/
- Run the clipper in Docker with GPU acceleration: https://dedupedai.com/use-cases/artbyjazi-autoclip/run-autoclip-in-docker-with-gpu/

## How to turn a long video into captioned vertical clips offline with Autoclip

Repo: [artbyjazi/autoclip](https://github.com/artbyjazi/autoclip) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README

Paste a YouTube link or drop a file and get ranked 9:16 clips with burned-in captions, all on your machine.

**Who it is for:** A podcaster or streamer who has a two hour recording and needs short clips for social feeds this week.

**The problem:** Finding the best moments and reframing them to vertical by hand takes hours per episode.

**The outcome:** Platform-ready MP4 clips, ranked by an LLM, reframed around the speaker with animated captions.

### Steps

1. Clone the repo with `git clone https://github.com/artbyjazi/autoclip.git`
2. Install with `cd autoclip && uv venv --python 3.11 && uv pip install -e ".[dev]"`
3. Build the UI with `cd frontend && npm install && npm run build && cd ..`
4. Pull a local model with `ollama pull llama3.1:8b` and check your setup with `autoclip doctor`
5. Start the app with `autoclip serve`, paste a link or drop a file, then review and export the clips

### Prompt to paste into your agent

```text
Install artbyjazi/autoclip, run `autoclip doctor` until it passes, then clip <path or YouTube URL> with the local Ollama provider. Report how many clips were exported and confirm each MP4 is 9:16 with captions burned in.
```

### Caveat

The README says it is pre-1.0 and the reframe quality bar has not yet been validated against a fixed golden set.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/artbyjazi-autoclip/clip-long-video-into-vertical-shorts/

## How to use your own LLM key for sharper clip selection with Autoclip

Repo: [artbyjazi/autoclip](https://github.com/artbyjazi/autoclip) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README

Point highlight detection at Anthropic, OpenAI, Gemini or an OpenAI-compatible endpoint while video stays local.

**Who it is for:** A video editor whose local model picks weak moments and who already pays for an API key.

**The problem:** Small local models can miss the moments that actually make good clips.

**The outcome:** Better ranked highlights, with only transcript text sent to the provider and never the video or audio.

### Steps

1. Install AutoClip and run `autoclip doctor`
2. Store your provider key with `autoclip config set-secret anthropic`
3. Start `autoclip serve` and choose the provider in the app
4. Run the same video again and compare the ranked clips

### Prompt to paste into your agent

```text
Configure artbyjazi/autoclip with `autoclip config set-secret anthropic`, re-run highlight detection on <video>, and list the top five clips with timestamps. Verify that no audio or video file was uploaded, only transcript text.
```

### Caveat

Only the transcript text leaves your machine with a hosted provider, so sensitive spoken content still reaches that provider.

Risk, credentials: Your provider API key is stored by AutoClip and billed per request, so use a key with a spending limit.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/artbyjazi-autoclip/better-clip-picks-with-your-own-key/

## How to run the clipper in Docker with GPU acceleration with Autoclip

Repo: [artbyjazi/autoclip](https://github.com/artbyjazi/autoclip) | Difficulty: Advanced | Time: 1 hour | Basis: From the README

Start AutoClip from the bundled compose file, optionally with the GPU profile for faster transcription.

**Who it is for:** A developer setting up a shared clipping box for a small content team.

**The problem:** Matching Python, MediaPipe and ffmpeg versions on every laptop is fragile.

**The outcome:** A containerized AutoClip that the team can reach from a browser.

### Steps

1. Clone the repo with `git clone https://github.com/artbyjazi/autoclip.git`
2. Start it with `docker compose -f docker/compose.yaml up --build`
3. On an NVIDIA machine use `docker compose -f docker/compose.yaml --profile gpu up --build`
4. Open the UI and run a test clip

### Prompt to paste into your agent

```text
Bring up artbyjazi/autoclip with `docker compose -f docker/compose.yaml --profile gpu up --build` on <host>, clip one short test video, and confirm the export finished and the GPU was used for transcription.
```

### Caveat

Transcription works on CPU but the README says a GPU speeds it up several-fold.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/artbyjazi-autoclip/run-autoclip-in-docker-with-gpu/