skainet-io-gguf/sk.ainet.io.gguf.llama/QuantizedTensorFactory/toQ4_K

toQ4_K

fun toQ4_K(rawTensor: Tensor<Int8, Byte>, logicalShape: Shape): Q4_KTensorData(source)

Convert a raw byte tensor to Q4_KTensorData.

Return

Q4_KTensorData ready for quantized matmul

Parameters

rawTensor

Tensor containing raw Q4_K bytes (loaded with RAW_BYTES policy)

logicalShape

The logical shape in elements (not bytes/blocks)

fun toQ4_K(rawTensor: Tensor<Int8, Byte>): Q4_KTensorData(source)

Convert a raw byte tensor to Q4_KTensorData using the tensor's shape.

Generated by Dokka
© 2026 Copyright