leideng/QCFuse / srt /layers /quantization /rocm_mxfp4_utils.py
leideng's picture
download
raw
327 Bytes
from aiter.ops.triton.batched_gemm_afp4wfp4_pre_quant import (
batched_gemm_afp4wfp4_pre_quant,
)
from aiter.ops.triton.fused_mxfp4_quant import (
fused_flatten_mxfp4_quant,
fused_rms_mxfp4_quant,
)
__all__ = [
"fused_rms_mxfp4_quant",
"fused_flatten_mxfp4_quant",
"batched_gemm_afp4wfp4_pre_quant",
]

Xet Storage Details

Size:
327 Bytes
·
Xet hash:
90dd422aebdd9ab65c47af3cd604b57e875320191831c535ee733a63a24e2c44

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.