hopper_tensor_core_bf16
Hopper Tensor Core BF16 GEMM prototype with a 64x64x16 WMMA tile and larger per-block warpgroup-style work.
This repository contains the standalone CUDA source for the hopper_tensor_core_bf16 lane from
the PyC kernel lab. It is a source artifact for inspection and benchmarking;
it is not a precompiled binary and the result below is not a universal ranking.
Performance
| Kernel | GPU / architecture | Shape | Best recorded result | Evidence |
|---|---|---|---|---|
hopper_tensor_core_bf16 |
sm90 | 1024x1024x1024 | 0.102 ms / 21.024 TFLOPS | Measured on sm90, shape 1024x1024x1024; evidence hopper-tensorcore-bringup-20260421T191817Z.json. |
The result is reported with the original campaign's timing and correctness context. Compare kernels only when GPU, CUDA version, matrix shape, warmup, repeats, and reference/correctness mode match.
Source
kernel.cu— copied fromkernels/prototypes/hopper/tensor_core/kernel.cu.- Original lane tags:
cuda, matmul, hopper, sm90, prototype, tensor-core, bf16.
Build/run contract
{nvcc} -O3 -std=c++17 -lineinfo -DPYC_HOPPER_TENSOR_CORE_USE_BF16=1 -gencode arch=compute_90,code=sm_90 -gencode arch=compute_90,code=compute_90 {source} -o {build_dir}/{name}
{build_dir}/{name} 1024 1024 1024 5 20
- Downloads last month
- -
cuda
kernel
gpu-optimization
hpc
apache-2.0