kernels-community
/

quantization

Model card Files Files and versions Community

quantization / fp8 /amd

Ctrl+K

Ctrl+K

3 contributors

History: 1 commit

danieldk's picture

danieldk HF Staff

Add `scaled_(int|fp8)_quant` and `fp8_marlin_gemm`

5c6fb68 5 months ago

hip_float8.h

3.9 kB

Add `scaled_(int|fp8)_quant` and `fp8_marlin_gemm` 5 months ago
hip_float8_impl.h

10.4 kB

Add `scaled_(int|fp8)_quant` and `fp8_marlin_gemm` 5 months ago
quant_utils.cuh

16.5 kB

Add `scaled_(int|fp8)_quant` and `fp8_marlin_gemm` 5 months ago