ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-06-15 05:18:07 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

sync : ggml (#2001)

Browse Source 
* sync : update scripts

* sync : ggml

* talk-llama : sync llama.cpp

* make : WHISPER_CUBLAS -> WHISPER_CUDA

* ci : try to fix sycl build

* talk-llama : fix make build
This commit is contained in:
Georgi Gerganov
2024-03-27 18:55:10 +02:00
committed by GitHub
parent 1558ec5a16
commit 2948c740a2
90 changed files with 15702 additions and 12476 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
ggml-common.h
View File

@ -377,6 +377,27 @@ typedef struct {
} block_iq1_s;
static_assert(sizeof(block_iq1_s) == sizeof(ggml_half) + QK_K/8 + QK_K/16, "wrong iq1_s block size/padding");
// 1.75 bpw
typedef struct {
uint8_t qs[QK_K/8]; // grid index, low 8 bits
uint8_t qh[QK_K/16]; // grid index, high 3 bits + grid shift bit (for two groups of 8)
#if QK_K == 64
ggml_half d;
#endif
uint8_t scales[QK_K/32]; // 3-bit block scales (4-bit if QK_K == 64)
} block_iq1_m;
#if QK_K == 64
static_assert(sizeof(block_iq1_m) == QK_K/8 + QK_K/16 + QK_K/32 + sizeof(ggml_half), "wrong iq1_m block size/padding");
#else
static_assert(sizeof(block_iq1_m) == QK_K/8 + QK_K/16 + QK_K/32, "wrong iq1_m block size/padding");
#endif
// Used by IQ1_M quants
typedef union {
ggml_half f16;
uint16_t u16;
} iq1m_scale_t;
// Non-linear quants
#define QK4_NL 32
typedef struct {
@ -1050,6 +1071,7 @@ GGML_TABLE_END()
#define NGRID_IQ1S 2048
#define IQ1S_DELTA 0.125f
#define IQ1M_DELTA 0.125f
#if defined(GGML_COMMON_IMPL_C)
GGML_TABLE_BEGIN(uint64_t, iq1s_grid, NGRID_IQ1S)
0xffffffffffffffff, 0xffffffffffffff01, 0xffffffffffff0000, 0xffffffffffff01ff,