[Quantization] Add ModelOpt IQ2_XS, IQ2_XXS, and Q8_0 support via b12x (#59604)
Signed-off-by: Andrii Skliar <askliar@nvidia.com> Signed-off-by: Luke Alonso <lalonso@gmail.com> Co-authored-by: Andrii Skliar <askliar@nvidia.com> Co-authored-by: Luke Alonso <lalonso@gmail.com>
This commit is contained in:
co-authored by
Andrii Skliar
Luke Alonso
parent
ac71b17517
commit
6849ca696a
@@ -11,6 +11,9 @@ from typing import Any
|
||||
|
||||
import torch
|
||||
|
||||
# Logical elements and encoded bytes per block.
|
||||
B12X_BLOCK_CODECS = {"iq2_xs": (256, 74), "iq2_xxs": (256, 66), "q8_0": (32, 34)}
|
||||
|
||||
|
||||
@dataclass(frozen=True)
|
||||
class B12xWarmupUnit:
|
||||
|
||||
Reference in New Issue
Block a user