Q8MemorySegmentTensorData
class Q8MemorySegmentTensorData(initialShape: Shape, val segment: MemorySegment, val segmentByteOffset: Long = 0) : TensorData<DType, Byte> , Q8MemorySegmentMarker(source)
Q8_0 quantized tensor data backed by a MemorySegment.
Q8_0 block layout (34 bytes per 32 elements):
2 bytes: f16 scale (little-endian)
32 bytes: int8 quantized codes
Dequantization: outputi = codei * scale
The segment is arena-managed and 64-byte aligned for SIMD access.
Properties
Link copied to clipboard
Number of Q8_0 blocks in the tensor.
Link copied to clipboard
Bytes per block (always 34).
Link copied to clipboard
Byte offset into segment where this tensor's data starts.