Source Package: clpeak (2.1.1-2.fc44)
Measure the peak achievable performance of GPU compute devices
A synthetic micro-benchmark that measures the peak achievable performance of GPU compute devices. It exercises tight vector / MAD / MMA loops and vendor-SDK GEMM libraries (cuBLASLt on NVIDIA, MPS on Apple) to expose what the hardware is capable of — from raw ALU peaks to near-vendor-advertised matrix throughput. clpeak began as an OpenCL-only tool. It now ships four interchangeable backends — OpenCL, Vulkan, CUDA, and Metal — running back-to-back on the same hardware, so cross-stack differences (driver lowering, instruction scheduling, extension exposure) become visible alongside the raw peak numbers.
Package information
- Homepage
- [https://github.com/krrishnarraj/clpeak]
- License
- Apache-2.0
- Mantenedor do pacote downstream
- filiperosset, rathannObserved Nov 20, 2026 [source]
The following binary packages are built from this source package:
- clpeak
Measure the peak achievable performance of GPU compute devices
