File size: 1,960 Bytes
2976eec | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 | from .quant import (
static_per_tensor_quant_fp8_i8,
dynamic_per_tensor_quant_fp8_i8,
dynamic_per_token_quant_fp8_i8,
dynamic_mxfp4_quant,
_mxfp4_quant_op,
dynamic_mxfp8_quant,
fp8_legacy_to_mxfp8,
_mxfp8_quant_op,
dynamic_nvfp4_quant,
_nvfp4_quant_op,
)
from .fused_fp8_quant import (
calc_rows_per_block,
fused_rms_fp8_per_tensor_static_quant,
fused_rms_fp8_group_quant,
fused_rms_gated_fp8_group_quant,
fused_flatten_fp8_group_quant,
fused_reduce_act_mul_fp8_group_quant,
fused_reduce_rms_fp8_group_quant,
get_fp8_min_max_bounds,
)
from .fused_mxfp4_quant import (
fused_rms_mxfp4_quant,
fused_flatten_mxfp4_quant,
fused_reduce_act_mul_and_mxfp4_quant,
fused_reduce_rms_mxfp4_quant,
fused_dynamic_mxfp4_quant_moe_sort,
)
from .fused_mxfp8_quant import (
fused_rms_mxfp8_quant,
fused_dual_rmsnorm_mxfp8_quant,
fused_flatten_mxfp8_quant,
)
__all__ = [
# quant.py exports
"static_per_tensor_quant_fp8_i8",
"dynamic_per_tensor_quant_fp8_i8",
"dynamic_per_token_quant_fp8_i8",
"dynamic_mxfp4_quant",
"_mxfp4_quant_op",
"dynamic_mxfp8_quant",
"fp8_legacy_to_mxfp8",
"_mxfp8_quant_op",
"dynamic_nvfp4_quant",
"_nvfp4_quant_op",
# fused_fp8_quant.py exports
"calc_rows_per_block",
"get_fp8_min_max_bounds",
"fused_rms_fp8_per_tensor_static_quant",
"fused_rms_fp8_group_quant",
"fused_rms_gated_fp8_group_quant",
"fused_flatten_fp8_group_quant",
"fused_reduce_act_mul_fp8_group_quant",
"fused_reduce_rms_fp8_group_quant",
# fused_mxfp4_quant.py exports
"fused_rms_mxfp4_quant",
"fused_flatten_mxfp4_quant",
"fused_reduce_act_mul_and_mxfp4_quant",
"fused_reduce_rms_mxfp4_quant",
"fused_dynamic_mxfp4_quant_moe_sort",
# fused_mxfp8_quant.py exports
"fused_rms_mxfp8_quant",
"fused_dual_rmsnorm_mxfp8_quant",
"fused_flatten_mxfp8_quant",
]
|