PyTorch model profiler for computing MACs and parameter counts to measure deep learning model complexity.
-
Updated
Oct 19, 2026 - Python
PyTorch model profiler for computing MACs and parameter counts to measure deep learning model complexity.
Memoraith: A lightweight, versatile model profiler for deep learning frameworks. Optimize your neural networks with detailed insights on memory usage, computation time, and performance bottlenecks.
Open-source PyTorch AI model profiler for latency, memory, energy (NVML/RAPL), and layer-wise bottlenecks. Measure green AI / ML efficiency no fabricated joules.
Hassle free profiling of ONNX models on various backends
To associate your repository with the model-profiling topic, visit your repo's landing page and select "manage topics."