Metadata-Version: 2.4
Name: nock-cli
Version: 0.1.0
Summary: Open models that live on your machine. Ask questions about your videos, offline.
Author: NockAI
License-Expression: Apache-2.0
Project-URL: Homepage, https://nockai.dev
Project-URL: Model, https://huggingface.co/Nock-AI/nock-video-4b-GGUF
Keywords: local,offline,llm,video,llama.cpp,gguf
Classifier: Environment :: Console
Classifier: Operating System :: Microsoft :: Windows
Classifier: Operating System :: POSIX :: Linux
Classifier: Operating System :: MacOS
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Multimedia :: Video
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Provides-Extra: links
Requires-Dist: yt-dlp; extra == "links"
Dynamic: license-file

# Nock CLI

Open models that live on your machine. Ask questions about your videos, offline.

```bash
nock video summarize my_video.mp4
nock video ask my_video.mp4 "What error message shows on screen?"
nock video find my_video.mp4 "when the error message shows up"
nock video summarize "https://www.youtube.com/watch?v=..."
```

Everything runs on your computer. Frames are read into a temporary folder, sent to a model running on
`127.0.0.1`, and deleted when the answer is ready. Nothing is uploaded.

## Install

You need Python 3.9 or newer and [ffmpeg](https://ffmpeg.org/download.html).

```bash
pipx install nock-cli
nock setup
nock doctor
```

`nock setup` downloads, once:
- the engine: an official [llama.cpp](https://github.com/ggml-org/llama.cpp) build (about 35 MB; Vulkan on Windows
  and Linux, Metal on Apple Silicon). Already have `llama-server`? Nock uses it.
- the model: [Nock Video 4B](https://huggingface.co/Nock-AI/nock-video-4b-GGUF) (about 3.4 GB).

Files go to `~/.nock` (set `NOCK_HOME` to use another folder).

### Links

Links are downloaded to your own machine with [yt-dlp](https://github.com/yt-dlp/yt-dlp), which you install
yourself:

```bash
pipx inject nock-cli yt-dlp
```

Only use videos you have the right to use.

## What to expect

Nock Video 4B is a **preview**. On our test of 20 videos, through llama.cpp at 64 frames, it found the moment in
42 of 56 questions, answered 56 of 60 questions about what is on screen, and gave 6 of 7 good summaries. Full
results, including the misses, are on the [model card](https://huggingface.co/Nock-AI/nock-video-4b-GGUF).

- It reads frames only, not sound.
- A long video is read as at most 64 frames, so short events in long videos can be missed.
- 64 frames used about 5.5 GB of GPU memory in our test. On GPUs under 6 GB, Nock uses 32 frames (finding the
  moment drops: 42 to 35 of 56 on our test). `--frames` overrides it.
- Without a GPU it works, but slowly: minutes per question.

## Options

| Option | Meaning |
|---|---|
| `--frames N` | frames to read (default 64, 32 on GPUs under 6 GB) |
| `nock setup --backend cpu` | CPU-only engine |
| `NOCK_LLAMA_SERVER=/path/llama-server` | use your own llama.cpp build |

Server logs are in `~/.nock/logs/` if something goes wrong.

## License

Apache-2.0. The model has its own license on its model card (Apache-2.0, from Qwen3.5 4B).
