philipturner / amx-benchmarks
Running linear algebra as fast as possible on Apple silicon
☆18Updated last year
Related projects: ⓘ
- A tool for generating information about the matrix multiplication instructions in AMD Radeon™ and AMD Instinct™ accelerators☆60Updated 8 months ago
- BLIS fork with kernels for Apple M1. (Perhaps) The first open-source BLAS with Apple Matrix Coprocessor support.☆33Updated last year
- Advanced Profiling and Analytics for AMD Hardware☆132Updated last week
- GPUOcelot: A dynamic compilation framework for PTX☆136Updated 3 months ago
- Study and Implementations of Numerical Algorithms on Apple M1 and A* Devices☆120Updated last year
- NVIDIA HPCG is based on the HPCG benchmark and optimized for performance on NVIDIA accelerated HPC systems.☆41Updated 3 weeks ago
- AMD lab notes with code examples to demonstrate use of AMD GPUs☆89Updated 2 months ago
- A framework that support executing unmodified CUDA source code on non-NVIDIA devices.☆94Updated last month
- Emulating double-precision arithmetic on Apple GPUs☆43Updated last year
- FP64 equivalent GEMM via Int8 Tensor Cores using the Ozaki scheme☆44Updated 2 weeks ago
- Test suite for probing the numerical behavior of NVIDIA tensor cores☆29Updated last month
- SYCL Benchmark Suite☆57Updated last week
- rocWMMA☆85Updated this week
- Benchmark for measuring the performance of sparse and irregular memory access.☆72Updated last week
- A simple profiler to count Nvidia PTX assembly instructions of OpenCL/SYCL/CUDA kernels for roofline model analysis.☆39Updated 8 months ago
- ROCm Parallel Primitives☆156Updated this week
- Tenstorrent MLIR compiler☆52Updated this week
- Trying to figure various CPU things out☆58Updated 7 months ago
- Tensor Tiling Library☆33Updated 3 weeks ago
- High Performance Linpack for Next-Generation AMD HPC Accelerators☆41Updated last week
- IREE's PyTorch Frontend, based on Torch Dynamo.☆43Updated this week
- Reusable software components for ROCm developers☆81Updated last week
- SYCL Reference Manual☆25Updated 4 months ago
- ☆15Updated 8 months ago
- ☆67Updated this week
- ☆124Updated this week
- ☆53Updated last week
- Next generation library for iterative sparse solvers for ROCm platform☆74Updated this week
- ☆42Updated last month
- CuPBoP-AMD is a CUDA translator that translates CUDA programs at NVVM IR level to HIP-compatible IR that can run on AMD GPUs.☆31Updated 10 months ago