Last released May 20, 2026
A lightweight PyTorch model profiler for latency, parameters, memory, and layer-wise bottlenecks.
Supported by