This is the repository card of drbh/fp8-weight-only-linear that has been pushed on the Hub. It was built to be used with the kernels library. This card was automatically generated.
How to use
# make sure `kernels` is installed: `pip install -U kernels`
from kernels import get_kernel
kernel_module = get_kernel("drbh/fp8-weight-only-linear")
ops = kernel_module.ops
ops(...)
Available functions
opsfp8_linear_w8a16quantize_weight
Benchmarks
No benchmark available yet.
Source code
Source code of this kernel originally comes from https://github.com/drbh/fp8-weight-only-linear and it was repurposed for compatibility with kernels.
- Downloads last month
- 4
mit
Supported hardwares new
CUDA
- OS
- linux
- Arch
- x86_64



