Metadata-Version: 2.5
Name: diescope
Version: 0.0.2
Summary: See what your silicon is actually doing. Chip-level utilization for GPUs, CPUs, and Apple silicon.
Project-URL: Homepage, https://diescope.com
Project-URL: Source, https://github.com/diescope/diescope
Author-email: Diescope <hi@diescope.dev>
License-Expression: Apache-2.0
License-File: LICENSE
Keywords: apple-silicon,cuda,gpu,observability,profiling,utilization
Classifier: Development Status :: 1 - Planning
Classifier: Environment :: Console
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: System Administrators
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: System :: Hardware
Classifier: Topic :: System :: Monitoring
Requires-Python: >=3.9
Description-Content-Type: text/markdown

# Diescope

See what your silicon is actually doing.

Standard tools report whether a chip is busy. Diescope reports how much of the chip is doing useful work: compute units active, tensor and matrix engines in use, and memory bandwidth consumed against the hardware's peak.

nvidia-smi can say 100% while most of the die sits idle. Diescope shows the gap.

## Status

Early development. This release reserves the name and installs a placeholder CLI. Measurement is not implemented yet.

## Planned support

- NVIDIA GPUs (NVML, DCGM, CUPTI)
- Apple silicon (powermetrics, Metal performance counters)
- AMD and Intel GPUs
- CPUs (perf, RAPL)

## Install

```
pip install diescope
cargo install diescope
```

## Links

- Website: https://diescope.com
- Source: https://github.com/diescope/diescope
- Contact: hi@diescope.dev
