Metadata-Version: 2.4
Name: cgpu-info
Version: 0.1.1
Summary: Quick CUDA/GPU status summary for ML engineers
Project-URL: Homepage, https://github.com/augustsletto/cgpu
Project-URL: Repository, https://github.com/augustsletto/cgpu
Project-URL: Issues, https://github.com/augustsletto/cgpu/issues
Author-email: August Sletto <augustsletto@gmail.com>
License-Expression: MIT
License-File: LICENSE
Keywords: cuda,deep-learning,gpu,machine-learning,nvidia,pytorch
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
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: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.8
Requires-Dist: colorama>=0.4.0
Provides-Extra: full
Requires-Dist: torch; extra == 'full'
Description-Content-Type: text/markdown

# cgpu

Quick CUDA/GPU status summary for ML engineers. One import, one call, all the info you need.

## Installation

```bash
pip install cgpu-info
```

or with uv:

```bash
uv pip install cgpu-info
```

**Note:** You need PyTorch installed separately. 

## Usage

```python
from cgpu import cgpu

device = cgpu()
# Now use `device` in your code
model.to(device)
```

That's it! You'll see a colorful summary like:

```
═══════════════════════════════════════
          GPU Status Summary
═══════════════════════════════════════
✓ CUDA Available
  Device: cuda
  GPU Count: 1
  [0] NVIDIA GeForce RTX 4090
      VRAM: 24.0 GB
      Allocated: 0.00 GB
      Reserved: 0.00 GB
      Temp: 42°C
      GPU Util: 0%
      Mem Util: 0%
  CUDA Version: 12.1
  cuDNN Version: 8902
  PyTorch: 2.1.0
═══════════════════════════════════════
```

## What it shows

- CUDA availability status
- Device string (`cuda` or `cpu`)
- GPU name and count
- VRAM total and usage
- Temperature (color-coded: green < 50°C, yellow < 70°C, red >= 70°C)
- GPU/Memory utilization
- CUDA, cuDNN, and PyTorch versions

## License

MIT
