11 lines
244 B
Python
11 lines
244 B
Python
"""GGML quantisation operations.
|
|
|
|
Provides numpy-based GGML block quantisation for architectures
|
|
not supported by llama.cpp.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
from helpers.ggml.quantiser import GGMLQuantiser
|
|
|
|
__all__ = ["GGMLQuantiser"]
|