Metadata-Version: 2.5
Name: z-vision-generator
Version: 0.14.0b1
Summary: Z-Vision Generator — cross-platform AI image and video generator
Project-URL: Homepage, https://github.com/knuthelge/ZVisionGenerator
Project-URL: Repository, https://github.com/knuthelge/ZVisionGenerator
Project-URL: Documentation, https://knuthelge.github.io/ZVisionGenerator/
Project-URL: Changelog, https://github.com/knuthelge/ZVisionGenerator/blob/main/CHANGELOG.md
Project-URL: Bug Tracker, https://github.com/knuthelge/ZVisionGenerator/issues
Author: Knut Risheim
License-Expression: AGPL-3.0-or-later
License-File: LICENSE
License-File: THIRD_PARTY_LICENSES.md
Keywords: ai,diffusers,flux,image-generation,mlx,stable-diffusion,video-generation,z-image
Classifier: Development Status :: 5 - Production/Stable
Classifier: Environment :: Console
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: End Users/Desktop
Classifier: License :: OSI Approved :: GNU Affero General Public License v3 or later (AGPLv3+)
Classifier: Operating System :: MacOS
Classifier: Operating System :: Microsoft :: Windows
Classifier: Operating System :: POSIX :: Linux
Classifier: Programming Language :: Python :: 3.14
Classifier: Topic :: Multimedia :: Graphics
Classifier: Topic :: Multimedia :: Video
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.14
Requires-Dist: accelerate>=1.15.0
Requires-Dist: av>=19.0.1; sys_platform != 'darwin'
Requires-Dist: bitsandbytes>=0.50.0; sys_platform != 'darwin'
Requires-Dist: diffusers==0.41.0
Requires-Dist: einops>=0.8.0; sys_platform != 'darwin'
Requires-Dist: fastapi>=0.143.0
Requires-Dist: huggingface-hub<2.0,>=1.33.0
Requires-Dist: mflux<0.23,>=0.22.0; sys_platform == 'darwin'
Requires-Dist: mlx-arsenal<0.3,>=0.2.4; sys_platform == 'darwin'
Requires-Dist: mlx-lm<0.33,>=0.32.0; sys_platform == 'darwin'
Requires-Dist: mlx-vlm==0.6.10; sys_platform == 'darwin'
Requires-Dist: mlx<0.33,>=0.32.2; sys_platform == 'darwin'
Requires-Dist: numpy>=2.0.0
Requires-Dist: omegaconf>=2.3.0; sys_platform != 'darwin'
Requires-Dist: opencv-python-headless>=4.10.0; sys_platform != 'darwin'
Requires-Dist: peft>=0.21.2; sys_platform != 'darwin'
Requires-Dist: pillow<13,>=12.3.0
Requires-Dist: psutil>=5.9.0
Requires-Dist: python-multipart>=0.0.32
Requires-Dist: pyyaml>=6.0.3
Requires-Dist: rotary-embedding-torch>=0.5.3; sys_platform != 'darwin'
Requires-Dist: ruamel-yaml>=0.18.10
Requires-Dist: safetensors>=0.8.0
Requires-Dist: scipy>=1.17.0
Requires-Dist: torch>=2.14.0
Requires-Dist: torchvision>=0.29.0; sys_platform != 'darwin'
Requires-Dist: tqdm>=4.66.0
Requires-Dist: transformers>=5.19.0
Requires-Dist: uvicorn>=0.54.0
Description-Content-Type: text/markdown

&nbsp;
<p align="center">
    <img src="https://raw.githubusercontent.com/knuthelge/ZVisionGenerator/main/docs/assets/ziv-logo.png" alt="Z-Vision Generator logo with the Blob mascot" width="360"/>
</p>
&nbsp;

# Z-Vision Generator

[![CI](https://github.com/knuthelge/ZVisionGenerator/actions/workflows/ci.yml/badge.svg)](https://github.com/knuthelge/ZVisionGenerator/actions/workflows/ci.yml)
[![PyPI](https://img.shields.io/pypi/v/z-vision-generator)](https://pypi.org/project/z-vision-generator/)
[![Python 3.14+](https://img.shields.io/badge/python-3.14%2B-blue)](https://www.python.org/downloads/)
[![License: AGPL-3.0](https://img.shields.io/badge/license-AGPL--3.0--or--later-blue)](LICENSE)

Local AI image and video generation — hassle-free and fun. No tangled node graphs, no cloud dependencies, just prompts and results. Runs on macOS (Apple Silicon / MLX) and on Windows and Linux with NVIDIA CUDA through diffusers.

## Features

- **Images** from a prompt, with Z-Image, FLUX.2 Klein, Krea 2 Turbo, Ideogram 4 Instant and, on Windows and Linux, MOSS-Image.
- **Videos with sound** from a prompt or a starting image, with LTX-2.3, LTX-2.5 and MiniMax H3. End on a chosen last frame with H3, and extend a clip to make it longer.
- **A prompt enhancer** that runs on your machine and turns a short idea into a detailed prompt.
- **Upscaling** of new and existing images, of video as it's made, and of existing videos on Windows and Linux.
- **Prompt files** with random choices, reusable snippets and batch runs.
- **LoRAs**, stacked and weighted, for images and video.
- **A Web UI** for your computer, phone or tablet, and a **CLI** for scripts and batches, sharing the same models, prompt files and outputs.

| Platform | Images | Video |
|---|---|---|
| macOS (Apple Silicon) | ✅ via MLX | ✅ via MLX |
| Windows (NVIDIA GPU) | ✅ via diffusers/CUDA | ✅ via diffusers/CUDA |
| Linux (NVIDIA GPU) | ✅ via diffusers/CUDA | ✅ via diffusers/CUDA |

## Install

Requires Python 3.14+, [uv](https://docs.astral.sh/uv/), and [ffmpeg](https://ffmpeg.org/) for video. On Windows and Linux: an RTX 20-series or newer NVIDIA GPU with driver 580 or newer.

```bash
uv tool install z-vision-generator
ziv ui
```

On Windows, add PyTorch's CUDA index to the install command; PyPI's PyTorch for Windows has no CUDA. See [Install](https://knuthelge.github.io/ZVisionGenerator/install/) for the command and the details.

The Web UI opens in your browser and offers to download an image model picked for your machine.

## Quick start

```bash
# Start the Web UI
ziv ui

# An image from the command line
ziv-image -m klein4b --prompt "A red fox sitting in fresh snow, early morning light"

# A batch from a prompt file, three runs each
ziv-image -m klein9b -p prompts.yaml -r 3

# A video with sound, and one from an image
ziv-video -m ltx-4 --prompt "A cat walking through a garden, birds singing"
ziv-video -m ltx-4 --image photo.jpg --prompt "The camera slowly zooms in"
```

## Documentation

**[knuthelge.github.io/ZVisionGenerator](https://knuthelge.github.io/ZVisionGenerator/)**

- [Web UI](https://knuthelge.github.io/ZVisionGenerator/web-ui/): your first image and video, and every page of the app
- [CLI](https://knuthelge.github.io/ZVisionGenerator/cli/): how-tos and the full command reference
- [Concepts](https://knuthelge.github.io/ZVisionGenerator/concepts/): models, prompts, the enhancer, upscaling, memory and configuration
- [Developer](https://knuthelge.github.io/ZVisionGenerator/developer/): setting up, conventions and architecture

## Contributing

Contributions are welcome. See [CONTRIBUTING.md](CONTRIBUTING.md) and the [developer docs](https://knuthelge.github.io/ZVisionGenerator/developer/).

## License

This project is licensed under the [GNU Affero General Public License v3.0 or later](LICENSE).
