Function quantize_nf4

bitsandbytes/functional.py:873–881

(
    A: torch.Tensor,
    absmax: Optional[torch.Tensor] = None,
    out: Optional[torch.Tensor] = None,
    blocksize=None,
    compress_statistics=False,
    quant_storage=torch.uint8,
)

def quantize_nf4(
    A: torch.Tensor,
    absmax: Optional[torch.Tensor] = None,
    out: Optional[torch.Tensor] = None,
    blocksize=None,
    compress_statistics=False,
    quant_storage=torch.uint8,
):
    return quantize_4bit(A, absmax, out, blocksize, compress_statistics, "nf4", quant_storage)


def quantize_4bit(