vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

ggml : use F16 instead of F32 in Q4_0, Q4_1, Q8_0 (#1508)

Browse source 
* ggml : use F16 instead of F32 in Q4_0, Q4_1 and Q8_0

* llama : bump LLAMA_FILE_VERSION to 3

* cuda : update Q4 and Q8 dequantize kernels

* ggml : fix AVX dot products

* readme : update performance table + hot topics
This commit is contained in:
Georgi Gerganov 2023-05-19 22:17:18 +03:00 • committed by GitHub
parent 6986c7835a
commit 2d5db48371
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
6 changed files with 109 additions and 102 deletions
Download patch file Download diff file Expand all files Collapse all files

14
ggml-cuda.cu
View file

@ -42,19 +42,19 @@ typedef void (*dequantize_mul_mat_vec_cuda_t)(const void * vx, const float * y,
#define QK4_0 32
#define QR4_0 2
typedef struct {
float d; // delta
half d; // delta
uint8_t qs[QK4_0 / 2]; // nibbles / quants
} block_q4_0;
static_assert(sizeof(block_q4_0) == sizeof(float) + QK4_0 / 2, "wrong q4_0 block size/padding");
static_assert(sizeof(block_q4_0) == sizeof(ggml_fp16_t) + QK4_0 / 2, "wrong q4_0 block size/padding");
#define QK4_1 32
#define QR4_1 2
typedef struct {
float d; // delta
float m; // min
half d; // delta
half m; // min
uint8_t qs[QK4_1 / 2]; // nibbles / quants
} block_q4_1;
static_assert(sizeof(block_q4_1) == sizeof(float) * 2 + QK4_1 / 2, "wrong q4_1 block size/padding");
static_assert(sizeof(block_q4_1) == sizeof(ggml_fp16_t) * 2 + QK4_1 / 2, "wrong q4_1 block size/padding");
#define QK5_0 32
#define QR5_0 2
@ -78,10 +78,10 @@ static_assert(sizeof(block_q5_1) == 2 * sizeof(ggml_fp16_t) + sizeof(uint32_t) +
#define QK8_0 32
#define QR8_0 1
typedef struct {
float d; // delta
half d; // delta
int8_t qs[QK8_0]; // quants
} block_q8_0;
static_assert(sizeof(block_q8_0) == sizeof(float) + QK8_0, "wrong q8_0 block size/padding");
static_assert(sizeof(block_q8_0) == sizeof(ggml_fp16_t) + QK8_0, "wrong q8_0 block size/padding");
#define CUDA_DEQUANTIZE_BLOCK_SIZE 256
#define CUDA_DMMV_BLOCK_SIZE 32 // dmmv = dequantize_mul_mat_vec