quantize_nvfp4: Quantize Data (NVFP4)

In ggmlR: 'GGML' Tensor Operations for Machine Learning

Quantize Data (NVFP4)

Description

Quantizes float data to NVFP4 format (NVIDIA FP4 with UE4M3 per-sub-block scale).

Usage

quantize_nvfp4(src_data, n_rows, n_per_row, imatrix = NULL)

Arguments

src_data	Numeric vector of float values to quantize
n_rows	Number of rows
n_per_row	Number of elements per row (must be multiple of 64)
imatrix	Optional importance matrix (currently ignored)

Value

Raw vector of quantized data

See Also

Other quantization: dequantize_row_iq2_xxs(), dequantize_row_mxfp4(), dequantize_row_nvfp4(), dequantize_row_q1_0(), dequantize_row_q2_K(), dequantize_row_q4_0(), dequantize_row_tq1_0(), ggml_quant_block_info(), iq2xs_free_impl(), iq2xs_init_impl(), iq3xs_free_impl(), iq3xs_init_impl(), quantize_iq2_xxs(), quantize_mxfp4(), quantize_q1_0(), quantize_q2_K(), quantize_q4_0(), quantize_row_iq3_xxs_ref(), quantize_row_mxfp4_ref(), quantize_row_q2_K_ref(), quantize_row_q4_0_ref(), quantize_row_tq1_0_ref(), quantize_tq1_0()

ggmlR documentation built on July 14, 2026, 1:08 a.m.