Metadata-Version: 2.4
Name: yolovault
Version: 0.2.0
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Operating System :: Microsoft :: Windows :: Windows 10
Classifier: Operating System :: Microsoft :: Windows :: Windows 11
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Programming Language :: Rust
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Summary: YOLOVault SDK: protected loading and native inference for YOLO models sold on YOLOModels
Keywords: yolo,object-detection,onnx,inference,licensing,drm
Author: YOLOModels
License: Proprietary
Requires-Python: >=3.8
Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
Project-URL: Documentation, https://yolomodels.pages.dev/docs
Project-URL: Homepage, https://yolomodels.pages.dev

# yolovault

Official Python SDK for [YOLOModels](https://yolomodels.pages.dev) — protected loading and
native inference for YOLO detection models.

```python
import yolovault
import cv2

model = yolovault.load(
    license_key="MOS-XXXX-XXXX-XXXX-XXXX",   # from your purchase
    model_path="model.enc",                   # downloaded from My Models
)

frame = cv2.imread("scene.jpg")[:, :, ::-1]   # RGB uint8 numpy array
for x1, y1, x2, y2, class_id, score in model.detect(frame):
    print(f"class {class_id}: {score:.2f} at ({x1:.0f},{y1:.0f})")

model.release()
```

- Encrypted models decrypt **in memory only** — plaintext weights never touch disk
- Inference runs **100% locally**: ONNX Runtime built in, TensorRT for `.engine` models (needs `pip install tensorrt` + NVIDIA GPU)
- `.onnx` / `.engine` = fully native protection; `.pt` = protected load
- Licenses are **machine-bound**; a background heartbeat keeps sessions alive with a 48h offline grace window

Full docs: https://yolomodels.pages.dev/docs

Supported: Windows x64 (Python 3.8+). `.onnx` works everywhere; `.engine` needs an NVIDIA GPU with TensorRT 10.x and CUDA (`pip install tensorrt`, then point `YV_TRT_LIB`/`YV_CUDART_LIB` at the pip `tensorrt_libs`/`nvidia\cuda_runtime\bin` folders if not on PATH). Linux/macOS wheels in progress.

