MCPcopy Create free account
hub / github.com/appdevforall/CodeOnTheGo / quantize_q8_0

Function quantize_q8_0

subprojects/llama.cpp/ggml/src/ggml-quants.c:2088–2093  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

2086}
2087
2088size_t quantize_q8_0(const float * GGML_RESTRICT src, void * GGML_RESTRICT dst, int64_t nrow, int64_t n_per_row, const float * quant_weights) {
2089 (void)quant_weights; // not used
2090 const size_t row_size = ggml_row_size(GGML_TYPE_Q8_0, n_per_row);
2091 quantize_row_q8_0_ref(src, dst, (int64_t)nrow*n_per_row);
2092 return nrow * row_size;
2093}
2094
2095size_t quantize_mxfp4(const float * GGML_RESTRICT src, void * GGML_RESTRICT dst, int64_t nrow, int64_t n_per_row, const float * quant_weights) {
2096 GGML_UNUSED(quant_weights);

Callers 1

ggml_quantize_chunkFunction · 0.85

Calls 2

ggml_row_sizeFunction · 0.85
quantize_row_q8_0_refFunction · 0.85

Tested by

no test coverage detected