ggml.h File Reference

#include <stdint.h>
#include <stddef.h>
#include <stdbool.h>

Go to the source code of this file.

Classes
struct	ggml_object
struct	ggml_tensor
struct	ggml_cgraph
struct	ggml_scratch
struct	ggml_init_params
struct	ggml_opt_params
struct	quantize_fns_t

Macros
#define	GGML_API
#define	GGML_FILE_MAGIC   0x67676d6c
#define	GGML_FILE_VERSION   1
#define	GGML_MAX_DIMS   4
#define	GGML_MAX_NODES   4096
#define	GGML_MAX_PARAMS   16
#define	GGML_MAX_CONTEXTS   64
#define	GGML_MAX_OPT   4
#define	GGML_DEFAULT_N_THREADS   4
#define	GGML_ASSERT(x)
#define	GGML_RESTRICT   restrict

Typedefs
typedef uint16_t	ggml_fp16_t
typedef void(*	ggml_unary_op_f32_t) (const int, float *, const float *)
typedef void(*	ggml_binary_op_f32_t) (const int, float *, const float *, const float *)
typedef void(*	dequantize_row_q_t) (const void *GGML_RESTRICT x, float *GGML_RESTRICT y, int k)
typedef void(*	quantize_row_q_t) (const float *GGML_RESTRICT x, void *GGML_RESTRICT y, int k)
typedef void(*	vec_dot_q_t) (const int n, float *GGML_RESTRICT s, const void *GGML_RESTRICT x, const void *GGML_RESTRICT y)

Enumerations
enum	ggml_type {   GGML_TYPE_F32 = 0 , GGML_TYPE_F16 = 1 , GGML_TYPE_Q4_0 = 2 , GGML_TYPE_Q4_1 = 3 ,   GGML_TYPE_Q4_2 = 4 , GGML_TYPE_Q5_0 = 6 , GGML_TYPE_Q5_1 = 7 , GGML_TYPE_Q8_0 = 8 ,   GGML_TYPE_Q8_1 = 9 , GGML_TYPE_I8 , GGML_TYPE_I16 , GGML_TYPE_I32 ,   GGML_TYPE_COUNT }
enum	ggml_ftype {   GGML_FTYPE_UNKNOWN = -1 , GGML_FTYPE_ALL_F32 = 0 , GGML_FTYPE_MOSTLY_F16 = 1 , GGML_FTYPE_MOSTLY_Q4_0 = 2 ,   GGML_FTYPE_MOSTLY_Q4_1 = 3 , GGML_FTYPE_MOSTLY_Q4_1_SOME_F16 = 4 , GGML_FTYPE_MOSTLY_Q4_2 = 5 , GGML_FTYPE_MOSTLY_Q8_0 = 7 ,   GGML_FTYPE_MOSTLY_Q5_0 = 8 , GGML_FTYPE_MOSTLY_Q5_1 = 9 }
enum	ggml_op {   GGML_OP_NONE = 0 , GGML_OP_DUP , GGML_OP_ADD , GGML_OP_SUB ,   GGML_OP_MUL , GGML_OP_DIV , GGML_OP_SQR , GGML_OP_SQRT ,   GGML_OP_SUM , GGML_OP_MEAN , GGML_OP_REPEAT , GGML_OP_ABS ,   GGML_OP_SGN , GGML_OP_NEG , GGML_OP_STEP , GGML_OP_RELU ,   GGML_OP_GELU , GGML_OP_SILU , GGML_OP_NORM , GGML_OP_RMS_NORM ,   GGML_OP_MUL_MAT , GGML_OP_SCALE , GGML_OP_CPY , GGML_OP_CONT ,   GGML_OP_RESHAPE , GGML_OP_VIEW , GGML_OP_PERMUTE , GGML_OP_TRANSPOSE ,   GGML_OP_GET_ROWS , GGML_OP_DIAG_MASK_INF , GGML_OP_SOFT_MAX , GGML_OP_ROPE ,   GGML_OP_ALIBI , GGML_OP_CONV_1D_1S , GGML_OP_CONV_1D_2S , GGML_OP_FLASH_ATTN ,   GGML_OP_FLASH_FF , GGML_OP_MAP_UNARY , GGML_OP_MAP_BINARY , GGML_OP_COUNT }
enum	ggml_opt_type { GGML_OPT_ADAM , GGML_OPT_LBFGS }
enum	ggml_linesearch { GGML_LINESEARCH_DEFAULT = 1 , GGML_LINESEARCH_BACKTRACKING_ARMIJO = 0 , GGML_LINESEARCH_BACKTRACKING_WOLFE = 1 , GGML_LINESEARCH_BACKTRACKING_STRONG_WOLFE = 2 }
enum	ggml_opt_result {   GGML_OPT_OK = 0 , GGML_OPT_DID_NOT_CONVERGE , GGML_OPT_NO_CONTEXT , GGML_OPT_INVALID_WOLFE ,   GGML_OPT_FAIL , GGML_LINESEARCH_FAIL = -128 , GGML_LINESEARCH_MINIMUM_STEP , GGML_LINESEARCH_MAXIMUM_STEP ,   GGML_LINESEARCH_MAXIMUM_ITERATIONS , GGML_LINESEARCH_INVALID_PARAMETERS }

Functions
GGML_API float	ggml_fp16_to_fp32 (ggml_fp16_t x)
GGML_API ggml_fp16_t	ggml_fp32_to_fp16 (float x)
GGML_API void	ggml_fp16_to_fp32_row (const ggml_fp16_t *x, float *y, size_t n)
GGML_API void	ggml_fp32_to_fp16_row (const float *x, ggml_fp16_t *y, size_t n)
GGML_API void	ggml_time_init (void)
GGML_API int64_t	ggml_time_ms (void)
GGML_API int64_t	ggml_time_us (void)
GGML_API int64_t	ggml_cycles (void)
GGML_API int64_t	ggml_cycles_per_ms (void)
GGML_API void	ggml_print_object (const struct ggml_object *obj)
GGML_API void	ggml_print_objects (const struct ggml_context *ctx)
GGML_API int64_t	ggml_nelements (const struct ggml_tensor *tensor)
GGML_API size_t	ggml_nbytes (const struct ggml_tensor *tensor)
GGML_API int	ggml_blck_size (enum ggml_type type)
GGML_API size_t	ggml_type_size (enum ggml_type type)
GGML_API float	ggml_type_sizef (enum ggml_type type)
GGML_API const char *	ggml_type_name (enum ggml_type type)
GGML_API size_t	ggml_element_size (const struct ggml_tensor *tensor)
GGML_API bool	ggml_is_quantized (enum ggml_type type)
GGML_API enum ggml_type	ggml_ftype_to_ggml_type (enum ggml_ftype ftype)
GGML_API struct ggml_context *	ggml_init (struct ggml_init_params params)
GGML_API void	ggml_free (struct ggml_context *ctx)
GGML_API size_t	ggml_used_mem (const struct ggml_context *ctx)
GGML_API size_t	ggml_set_scratch (struct ggml_context *ctx, struct ggml_scratch scratch)
GGML_API struct ggml_tensor *	ggml_new_tensor (struct ggml_context *ctx, enum ggml_type type, int n_dims, const int64_t *ne)
GGML_API struct ggml_tensor *	ggml_new_tensor_1d (struct ggml_context *ctx, enum ggml_type type, int64_t ne0)
GGML_API struct ggml_tensor *	ggml_new_tensor_2d (struct ggml_context *ctx, enum ggml_type type, int64_t ne0, int64_t ne1)
GGML_API struct ggml_tensor *	ggml_new_tensor_3d (struct ggml_context *ctx, enum ggml_type type, int64_t ne0, int64_t ne1, int64_t ne2)
GGML_API struct ggml_tensor *	ggml_new_tensor_4d (struct ggml_context *ctx, enum ggml_type type, int64_t ne0, int64_t ne1, int64_t ne2, int64_t ne3)
GGML_API struct ggml_tensor *	ggml_new_i32 (struct ggml_context *ctx, int32_t value)
GGML_API struct ggml_tensor *	ggml_new_f32 (struct ggml_context *ctx, float value)
GGML_API struct ggml_tensor *	ggml_dup_tensor (struct ggml_context *ctx, const struct ggml_tensor *src)
GGML_API struct ggml_tensor *	ggml_view_tensor (struct ggml_context *ctx, const struct ggml_tensor *src)
GGML_API struct ggml_tensor *	ggml_set_zero (struct ggml_tensor *tensor)
GGML_API struct ggml_tensor *	ggml_set_i32 (struct ggml_tensor *tensor, int32_t value)
GGML_API struct ggml_tensor *	ggml_set_f32 (struct ggml_tensor *tensor, float value)
GGML_API int32_t	ggml_get_i32_1d (const struct ggml_tensor *tensor, int i)
GGML_API void	ggml_set_i32_1d (const struct ggml_tensor *tensor, int i, int32_t value)
GGML_API float	ggml_get_f32_1d (const struct ggml_tensor *tensor, int i)
GGML_API void	ggml_set_f32_1d (const struct ggml_tensor *tensor, int i, float value)
GGML_API void *	ggml_get_data (const struct ggml_tensor *tensor)
GGML_API float *	ggml_get_data_f32 (const struct ggml_tensor *tensor)
GGML_API const char *	ggml_get_name (const struct ggml_tensor *tensor)
GGML_API void	ggml_set_name (struct ggml_tensor *tensor, const char *name)
GGML_API struct ggml_tensor *	ggml_dup (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_add (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_add_inplace (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_sub (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_mul (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_div (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_sqr (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_sqrt (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_sum (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_mean (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_repeat (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_abs (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_sgn (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_neg (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_step (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_relu (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_gelu (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_silu (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_norm (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_rms_norm (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_mul_mat (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_scale (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_cpy (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_cont (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_reshape (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_reshape_2d (struct ggml_context *ctx, struct ggml_tensor *a, int64_t ne0, int64_t ne1)
GGML_API struct ggml_tensor *	ggml_reshape_3d (struct ggml_context *ctx, struct ggml_tensor *a, int64_t ne0, int64_t ne1, int64_t ne2)
GGML_API struct ggml_tensor *	ggml_view_1d (struct ggml_context *ctx, struct ggml_tensor *a, int64_t ne0, size_t offset)
GGML_API struct ggml_tensor *	ggml_view_2d (struct ggml_context *ctx, struct ggml_tensor *a, int64_t ne0, int64_t ne1, size_t nb1, size_t offset)
GGML_API struct ggml_tensor *	ggml_view_3d (struct ggml_context *ctx, struct ggml_tensor *a, int64_t ne0, int64_t ne1, int64_t ne2, size_t nb1, size_t nb2, size_t offset)
GGML_API struct ggml_tensor *	ggml_permute (struct ggml_context *ctx, struct ggml_tensor *a, int axis0, int axis1, int axis2, int axis3)
GGML_API struct ggml_tensor *	ggml_transpose (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_get_rows (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_diag_mask_inf (struct ggml_context *ctx, struct ggml_tensor *a, int n_past)
GGML_API struct ggml_tensor *	ggml_soft_max (struct ggml_context *ctx, struct ggml_tensor *a)
GGML_API struct ggml_tensor *	ggml_rope (struct ggml_context *ctx, struct ggml_tensor *a, int n_past, int n_dims, int mode)
struct ggml_tensor *	ggml_alibi (struct ggml_context *ctx, struct ggml_tensor *a, int n_past, int n_head)
GGML_API struct ggml_tensor *	ggml_conv_1d_1s (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_conv_1d_2s (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b)
GGML_API struct ggml_tensor *	ggml_flash_attn (struct ggml_context *ctx, struct ggml_tensor *q, struct ggml_tensor *k, struct ggml_tensor *v, bool masked)
GGML_API struct ggml_tensor *	ggml_flash_ff (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b0, struct ggml_tensor *b1, struct ggml_tensor *c0, struct ggml_tensor *c1)
GGML_API struct ggml_tensor *	ggml_map_unary_f32 (struct ggml_context *ctx, struct ggml_tensor *a, const ggml_unary_op_f32_t fun)
GGML_API struct ggml_tensor *	ggml_map_binary_f32 (struct ggml_context *ctx, struct ggml_tensor *a, struct ggml_tensor *b, const ggml_binary_op_f32_t fun)
GGML_API void	ggml_set_param (struct ggml_context *ctx, struct ggml_tensor *tensor)
GGML_API void	ggml_build_forward_expand (struct ggml_cgraph *cgraph, struct ggml_tensor *tensor)
GGML_API struct ggml_cgraph	ggml_build_forward (struct ggml_tensor *tensor)
GGML_API struct ggml_cgraph	ggml_build_backward (struct ggml_context *ctx, struct ggml_cgraph *gf, bool keep)
GGML_API void	ggml_graph_compute (struct ggml_context *ctx, struct ggml_cgraph *cgraph)
GGML_API void	ggml_graph_reset (struct ggml_cgraph *cgraph)
GGML_API void	ggml_graph_print (const struct ggml_cgraph *cgraph)
GGML_API void	ggml_graph_dump_dot (const struct ggml_cgraph *gb, const struct ggml_cgraph *gf, const char *filename)
GGML_API struct ggml_opt_params	ggml_opt_default_params (enum ggml_opt_type type)
GGML_API enum ggml_opt_result	ggml_opt (struct ggml_context *ctx, struct ggml_opt_params params, struct ggml_tensor *f)
GGML_API size_t	ggml_quantize_q4_0 (const float *src, void *dst, int n, int k, int64_t *hist)
GGML_API size_t	ggml_quantize_q4_1 (const float *src, void *dst, int n, int k, int64_t *hist)
GGML_API size_t	ggml_quantize_q4_2 (const float *src, void *dst, int n, int k, int64_t *hist)
GGML_API size_t	ggml_quantize_q5_0 (const float *src, void *dst, int n, int k, int64_t *hist)
GGML_API size_t	ggml_quantize_q5_1 (const float *src, void *dst, int n, int k, int64_t *hist)
GGML_API size_t	ggml_quantize_q8_0 (const float *src, void *dst, int n, int k, int64_t *hist)
GGML_API size_t	ggml_quantize_chunk (enum ggml_type type, const float *src, void *dst, int start, int n, int64_t *hist)
GGML_API int	ggml_cpu_has_avx (void)
GGML_API int	ggml_cpu_has_avx2 (void)
GGML_API int	ggml_cpu_has_avx512 (void)
GGML_API int	ggml_cpu_has_avx512_vbmi (void)
GGML_API int	ggml_cpu_has_avx512_vnni (void)
GGML_API int	ggml_cpu_has_fma (void)
GGML_API int	ggml_cpu_has_neon (void)
GGML_API int	ggml_cpu_has_arm_fma (void)
GGML_API int	ggml_cpu_has_f16c (void)
GGML_API int	ggml_cpu_has_fp16_va (void)
GGML_API int	ggml_cpu_has_wasm_simd (void)
GGML_API int	ggml_cpu_has_blas (void)
GGML_API int	ggml_cpu_has_cublas (void)
GGML_API int	ggml_cpu_has_clblast (void)
GGML_API int	ggml_cpu_has_gpublas (void)
GGML_API int	ggml_cpu_has_sse3 (void)
GGML_API int	ggml_cpu_has_vsx (void)
quantize_fns_t	ggml_internal_get_quantize_fn (size_t i)

Variables
static const size_t	GGML_OBJECT_SIZE = sizeof(struct ggml_object)

Macro Definition Documentation

GGML_API

#define GGML_API

GGML_ASSERT

#define GGML_ASSERT

(

x

)

Value:

    do                                                                                                                 \
    {                                                                                                                  \
        if (!(x))                                                                                                      \
        {                                                                                                              \
            fprintf(stderr, "GGML_ASSERT: %s:%d: %s\n", __FILE__, __LINE__, #x);                                       \
            abort();                                                                                                   \
        }                                                                                                              \
    } while (0)

GGML_DEFAULT_N_THREADS

#define GGML_DEFAULT_N_THREADS   4

GGML_FILE_MAGIC

#define GGML_FILE_MAGIC   0x67676d6c

GGML_FILE_VERSION

#define GGML_FILE_VERSION   1

GGML_MAX_CONTEXTS

#define GGML_MAX_CONTEXTS   64

GGML_MAX_DIMS

#define GGML_MAX_DIMS   4

GGML_MAX_NODES

#define GGML_MAX_NODES   4096

GGML_MAX_OPT

#define GGML_MAX_OPT   4

GGML_MAX_PARAMS

#define GGML_MAX_PARAMS   16

GGML_RESTRICT

#define GGML_RESTRICT   restrict

Typedef Documentation

dequantize_row_q_t

typedef void(* dequantize_row_q_t) (const void *GGML_RESTRICT x, float *GGML_RESTRICT y, int k)

ggml_binary_op_f32_t

typedef void(* ggml_binary_op_f32_t) (const int, float *, const float *, const float *)

ggml_fp16_t

typedef uint16_t ggml_fp16_t

ggml_unary_op_f32_t

typedef void(* ggml_unary_op_f32_t) (const int, float *, const float *)

quantize_row_q_t

typedef void(* quantize_row_q_t) (const float *GGML_RESTRICT x, void *GGML_RESTRICT y, int k)

vec_dot_q_t

typedef void(* vec_dot_q_t) (const int n, float *GGML_RESTRICT s, const void *GGML_RESTRICT x, const void *GGML_RESTRICT y)

Enumeration Type Documentation

ggml_ftype

enum ggml_ftype

Enumerator
GGML_FTYPE_UNKNOWN
GGML_FTYPE_ALL_F32
GGML_FTYPE_MOSTLY_F16
GGML_FTYPE_MOSTLY_Q4_0
GGML_FTYPE_MOSTLY_Q4_1
GGML_FTYPE_MOSTLY_Q4_1_SOME_F16
GGML_FTYPE_MOSTLY_Q4_2
GGML_FTYPE_MOSTLY_Q8_0
GGML_FTYPE_MOSTLY_Q5_0
GGML_FTYPE_MOSTLY_Q5_1

ggml_linesearch

enum ggml_linesearch

Enumerator
GGML_LINESEARCH_DEFAULT
GGML_LINESEARCH_BACKTRACKING_ARMIJO
GGML_LINESEARCH_BACKTRACKING_WOLFE
GGML_LINESEARCH_BACKTRACKING_STRONG_WOLFE

ggml_op

enum ggml_op

Enumerator
GGML_OP_NONE
GGML_OP_DUP
GGML_OP_ADD
GGML_OP_SUB
GGML_OP_MUL
GGML_OP_DIV
GGML_OP_SQR
GGML_OP_SQRT
GGML_OP_SUM
GGML_OP_MEAN
GGML_OP_REPEAT
GGML_OP_ABS
GGML_OP_SGN
GGML_OP_NEG
GGML_OP_STEP
GGML_OP_RELU
GGML_OP_GELU
GGML_OP_SILU
GGML_OP_NORM
GGML_OP_RMS_NORM
GGML_OP_MUL_MAT
GGML_OP_SCALE
GGML_OP_CPY
GGML_OP_CONT
GGML_OP_RESHAPE
GGML_OP_VIEW
GGML_OP_PERMUTE
GGML_OP_TRANSPOSE
GGML_OP_GET_ROWS
GGML_OP_DIAG_MASK_INF
GGML_OP_SOFT_MAX
GGML_OP_ROPE
GGML_OP_ALIBI
GGML_OP_CONV_1D_1S
GGML_OP_CONV_1D_2S
GGML_OP_FLASH_ATTN
GGML_OP_FLASH_FF
GGML_OP_MAP_UNARY
GGML_OP_MAP_BINARY
GGML_OP_COUNT

ggml_opt_result

enum ggml_opt_result

Enumerator
GGML_OPT_OK
GGML_OPT_DID_NOT_CONVERGE
GGML_OPT_NO_CONTEXT
GGML_OPT_INVALID_WOLFE
GGML_OPT_FAIL
GGML_LINESEARCH_FAIL
GGML_LINESEARCH_MINIMUM_STEP
GGML_LINESEARCH_MAXIMUM_STEP
GGML_LINESEARCH_MAXIMUM_ITERATIONS
GGML_LINESEARCH_INVALID_PARAMETERS

ggml_opt_type

enum ggml_opt_type

Enumerator
GGML_OPT_ADAM
GGML_OPT_LBFGS

ggml_type

enum ggml_type

Enumerator
GGML_TYPE_F32
GGML_TYPE_F16
GGML_TYPE_Q4_0
GGML_TYPE_Q4_1
GGML_TYPE_Q4_2
GGML_TYPE_Q5_0
GGML_TYPE_Q5_1
GGML_TYPE_Q8_0
GGML_TYPE_Q8_1
GGML_TYPE_I8
GGML_TYPE_I16
GGML_TYPE_I32
GGML_TYPE_COUNT

Function Documentation

ggml_abs()

GGML_API struct ggml_tensor * ggml_abs	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_add()

GGML_API struct ggml_tensor * ggml_add	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_add_inplace()

GGML_API struct ggml_tensor * ggml_add_inplace	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_alibi()

struct ggml_tensor * ggml_alibi	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		int	n_past,
		int	n_head )

ggml_blck_size()

GGML_API int ggml_blck_size

(

enum ggml_type

type

)

ggml_build_backward()

GGML_API struct ggml_cgraph ggml_build_backward	(	struct ggml_context *	ctx,
		struct ggml_cgraph *	gf,
		bool	keep )

ggml_build_forward()

GGML_API struct ggml_cgraph ggml_build_forward

(

struct ggml_tensor *

tensor

)

ggml_build_forward_expand()

GGML_API void ggml_build_forward_expand	(	struct ggml_cgraph *	cgraph,
		struct ggml_tensor *	tensor )

ggml_cont()

GGML_API struct ggml_tensor * ggml_cont	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_conv_1d_1s()

GGML_API struct ggml_tensor * ggml_conv_1d_1s	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_conv_1d_2s()

GGML_API struct ggml_tensor * ggml_conv_1d_2s	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_cpu_has_arm_fma()

GGML_API int ggml_cpu_has_arm_fma

(

void

)

ggml_cpu_has_avx()

GGML_API int ggml_cpu_has_avx

(

void

)

ggml_cpu_has_avx2()

GGML_API int ggml_cpu_has_avx2

(

void

)

ggml_cpu_has_avx512()

GGML_API int ggml_cpu_has_avx512

(

void

)

ggml_cpu_has_avx512_vbmi()

GGML_API int ggml_cpu_has_avx512_vbmi

(

void

)

ggml_cpu_has_avx512_vnni()

GGML_API int ggml_cpu_has_avx512_vnni

(

void

)

ggml_cpu_has_blas()

GGML_API int ggml_cpu_has_blas

(

void

)

ggml_cpu_has_clblast()

GGML_API int ggml_cpu_has_clblast

(

void

)

ggml_cpu_has_cublas()

GGML_API int ggml_cpu_has_cublas

(

void

)

ggml_cpu_has_f16c()

GGML_API int ggml_cpu_has_f16c

(

void

)

ggml_cpu_has_fma()

GGML_API int ggml_cpu_has_fma

(

void

)

ggml_cpu_has_fp16_va()

GGML_API int ggml_cpu_has_fp16_va

(

void

)

ggml_cpu_has_gpublas()

GGML_API int ggml_cpu_has_gpublas

(

void

)

ggml_cpu_has_neon()

GGML_API int ggml_cpu_has_neon

(

void

)

ggml_cpu_has_sse3()

GGML_API int ggml_cpu_has_sse3

(

void

)

ggml_cpu_has_vsx()

GGML_API int ggml_cpu_has_vsx

(

void

)

ggml_cpu_has_wasm_simd()

GGML_API int ggml_cpu_has_wasm_simd

(

void

)

ggml_cpy()

GGML_API struct ggml_tensor * ggml_cpy	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_cycles()

GGML_API int64_t ggml_cycles

(

void

)

ggml_cycles_per_ms()

GGML_API int64_t ggml_cycles_per_ms

(

void

)

ggml_diag_mask_inf()

GGML_API struct ggml_tensor * ggml_diag_mask_inf	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		int	n_past )

ggml_div()

GGML_API struct ggml_tensor * ggml_div	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_dup()

GGML_API struct ggml_tensor * ggml_dup	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_dup_tensor()

GGML_API struct ggml_tensor * ggml_dup_tensor	(	struct ggml_context *	ctx,
		const struct ggml_tensor *	src )

ggml_element_size()

GGML_API size_t ggml_element_size

(

const struct ggml_tensor *

tensor

)

ggml_flash_attn()

GGML_API struct ggml_tensor * ggml_flash_attn	(	struct ggml_context *	ctx,
		struct ggml_tensor *	q,
		struct ggml_tensor *	k,
		struct ggml_tensor *	v,
		bool	masked )

ggml_flash_ff()

GGML_API struct ggml_tensor * ggml_flash_ff	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b0,
		struct ggml_tensor *	b1,
		struct ggml_tensor *	c0,
		struct ggml_tensor *	c1 )

ggml_fp16_to_fp32()

GGML_API float ggml_fp16_to_fp32

(

ggml_fp16_t

x

)

ggml_fp16_to_fp32_row()

GGML_API void ggml_fp16_to_fp32_row	(	const ggml_fp16_t *	x,
		float *	y,
		size_t	n )

ggml_fp32_to_fp16()

GGML_API ggml_fp16_t ggml_fp32_to_fp16

(

float

x

)

ggml_fp32_to_fp16_row()

GGML_API void ggml_fp32_to_fp16_row	(	const float *	x,
		ggml_fp16_t *	y,
		size_t	n )

ggml_free()

GGML_API void ggml_free

(

struct ggml_context *

ctx

)

ggml_ftype_to_ggml_type()

GGML_API enum ggml_type ggml_ftype_to_ggml_type

(

enum ggml_ftype

ftype

)

ggml_gelu()

GGML_API struct ggml_tensor * ggml_gelu	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_get_data()

GGML_API void * ggml_get_data

(

const struct ggml_tensor *

tensor

)

ggml_get_data_f32()

GGML_API float * ggml_get_data_f32

(

const struct ggml_tensor *

tensor

)

ggml_get_f32_1d()

GGML_API float ggml_get_f32_1d	(	const struct ggml_tensor *	tensor,
		int	i )

ggml_get_i32_1d()

GGML_API int32_t ggml_get_i32_1d	(	const struct ggml_tensor *	tensor,
		int	i )

ggml_get_name()

GGML_API const char * ggml_get_name

(

const struct ggml_tensor *

tensor

)

ggml_get_rows()

GGML_API struct ggml_tensor * ggml_get_rows	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_graph_compute()

GGML_API void ggml_graph_compute	(	struct ggml_context *	ctx,
		struct ggml_cgraph *	cgraph )

ggml_graph_dump_dot()

GGML_API void ggml_graph_dump_dot	(	const struct ggml_cgraph *	gb,
		const struct ggml_cgraph *	gf,
		const char *	filename )

ggml_graph_print()

GGML_API void ggml_graph_print

(

const struct ggml_cgraph *

cgraph

)

ggml_graph_reset()

GGML_API void ggml_graph_reset

(

struct ggml_cgraph *

cgraph

)

ggml_init()

GGML_API struct ggml_context * ggml_init

(

struct ggml_init_params

params

)

ggml_internal_get_quantize_fn()

quantize_fns_t ggml_internal_get_quantize_fn

(

size_t

i

)

ggml_is_quantized()

GGML_API bool ggml_is_quantized

(

enum ggml_type

type

)

ggml_map_binary_f32()

GGML_API struct ggml_tensor * ggml_map_binary_f32	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b,
		const ggml_binary_op_f32_t	fun )

ggml_map_unary_f32()

GGML_API struct ggml_tensor * ggml_map_unary_f32	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		const ggml_unary_op_f32_t	fun )

ggml_mean()

GGML_API struct ggml_tensor * ggml_mean	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_mul()

GGML_API struct ggml_tensor * ggml_mul	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_mul_mat()

GGML_API struct ggml_tensor * ggml_mul_mat	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_nbytes()

GGML_API size_t ggml_nbytes

(

const struct ggml_tensor *

tensor

)

ggml_neg()

GGML_API struct ggml_tensor * ggml_neg	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_nelements()

GGML_API int64_t ggml_nelements

(

const struct ggml_tensor *

tensor

)

ggml_new_f32()

GGML_API struct ggml_tensor * ggml_new_f32	(	struct ggml_context *	ctx,
		float	value )

ggml_new_i32()

GGML_API struct ggml_tensor * ggml_new_i32	(	struct ggml_context *	ctx,
		int32_t	value )

ggml_new_tensor()

GGML_API struct ggml_tensor * ggml_new_tensor	(	struct ggml_context *	ctx,
		enum ggml_type	type,
		int	n_dims,
		const int64_t *	ne )

ggml_new_tensor_1d()

GGML_API struct ggml_tensor * ggml_new_tensor_1d	(	struct ggml_context *	ctx,
		enum ggml_type	type,
		int64_t	ne0 )

ggml_new_tensor_2d()

GGML_API struct ggml_tensor * ggml_new_tensor_2d	(	struct ggml_context *	ctx,
		enum ggml_type	type,
		int64_t	ne0,
		int64_t	ne1 )

ggml_new_tensor_3d()

GGML_API struct ggml_tensor * ggml_new_tensor_3d	(	struct ggml_context *	ctx,
		enum ggml_type	type,
		int64_t	ne0,
		int64_t	ne1,
		int64_t	ne2 )

ggml_new_tensor_4d()

GGML_API struct ggml_tensor * ggml_new_tensor_4d	(	struct ggml_context *	ctx,
		enum ggml_type	type,
		int64_t	ne0,
		int64_t	ne1,
		int64_t	ne2,
		int64_t	ne3 )

ggml_norm()

GGML_API struct ggml_tensor * ggml_norm	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_opt()

GGML_API enum ggml_opt_result ggml_opt	(	struct ggml_context *	ctx,
		struct ggml_opt_params	params,
		struct ggml_tensor *	f )

ggml_opt_default_params()

GGML_API struct ggml_opt_params ggml_opt_default_params

(

enum ggml_opt_type

type

)

ggml_permute()

GGML_API struct ggml_tensor * ggml_permute	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		int	axis0,
		int	axis1,
		int	axis2,
		int	axis3 )

ggml_print_object()

GGML_API void ggml_print_object

(

const struct ggml_object *

obj

)

ggml_print_objects()

GGML_API void ggml_print_objects

(

const struct ggml_context *

ctx

)

ggml_quantize_chunk()

GGML_API size_t ggml_quantize_chunk	(	enum ggml_type	type,
		const float *	src,
		void *	dst,
		int	start,
		int	n,
		int64_t *	hist )

ggml_quantize_q4_0()

GGML_API size_t ggml_quantize_q4_0	(	const float *	src,
		void *	dst,
		int	n,
		int	k,
		int64_t *	hist )

ggml_quantize_q4_1()

GGML_API size_t ggml_quantize_q4_1	(	const float *	src,
		void *	dst,
		int	n,
		int	k,
		int64_t *	hist )

ggml_quantize_q4_2()

GGML_API size_t ggml_quantize_q4_2	(	const float *	src,
		void *	dst,
		int	n,
		int	k,
		int64_t *	hist )

ggml_quantize_q5_0()

GGML_API size_t ggml_quantize_q5_0	(	const float *	src,
		void *	dst,
		int	n,
		int	k,
		int64_t *	hist )

ggml_quantize_q5_1()

GGML_API size_t ggml_quantize_q5_1	(	const float *	src,
		void *	dst,
		int	n,
		int	k,
		int64_t *	hist )

ggml_quantize_q8_0()

GGML_API size_t ggml_quantize_q8_0	(	const float *	src,
		void *	dst,
		int	n,
		int	k,
		int64_t *	hist )

ggml_relu()

GGML_API struct ggml_tensor * ggml_relu	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_repeat()

GGML_API struct ggml_tensor * ggml_repeat	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_reshape()

GGML_API struct ggml_tensor * ggml_reshape	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_reshape_2d()

GGML_API struct ggml_tensor * ggml_reshape_2d	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		int64_t	ne0,
		int64_t	ne1 )

ggml_reshape_3d()

GGML_API struct ggml_tensor * ggml_reshape_3d	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		int64_t	ne0,
		int64_t	ne1,
		int64_t	ne2 )

ggml_rms_norm()

GGML_API struct ggml_tensor * ggml_rms_norm	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_rope()

GGML_API struct ggml_tensor * ggml_rope	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		int	n_past,
		int	n_dims,
		int	mode )

ggml_scale()

GGML_API struct ggml_tensor * ggml_scale	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_set_f32()

GGML_API struct ggml_tensor * ggml_set_f32	(	struct ggml_tensor *	tensor,
		float	value )

ggml_set_f32_1d()

GGML_API void ggml_set_f32_1d	(	const struct ggml_tensor *	tensor,
		int	i,
		float	value )

ggml_set_i32()

GGML_API struct ggml_tensor * ggml_set_i32	(	struct ggml_tensor *	tensor,
		int32_t	value )

ggml_set_i32_1d()

GGML_API void ggml_set_i32_1d	(	const struct ggml_tensor *	tensor,
		int	i,
		int32_t	value )

ggml_set_name()

GGML_API void ggml_set_name	(	struct ggml_tensor *	tensor,
		const char *	name )

ggml_set_param()

GGML_API void ggml_set_param	(	struct ggml_context *	ctx,
		struct ggml_tensor *	tensor )

ggml_set_scratch()

GGML_API size_t ggml_set_scratch	(	struct ggml_context *	ctx,
		struct ggml_scratch	scratch )

ggml_set_zero()

GGML_API struct ggml_tensor * ggml_set_zero

(

struct ggml_tensor *

tensor

)

ggml_sgn()

GGML_API struct ggml_tensor * ggml_sgn	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_silu()

GGML_API struct ggml_tensor * ggml_silu	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_soft_max()

GGML_API struct ggml_tensor * ggml_soft_max	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_sqr()

GGML_API struct ggml_tensor * ggml_sqr	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_sqrt()

GGML_API struct ggml_tensor * ggml_sqrt	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_step()

GGML_API struct ggml_tensor * ggml_step	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_sub()

GGML_API struct ggml_tensor * ggml_sub	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		struct ggml_tensor *	b )

ggml_sum()

GGML_API struct ggml_tensor * ggml_sum	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_time_init()

GGML_API void ggml_time_init

(

void

)

ggml_time_ms()

GGML_API int64_t ggml_time_ms

(

void

)

ggml_time_us()

GGML_API int64_t ggml_time_us

(

void

)

ggml_transpose()

GGML_API struct ggml_tensor * ggml_transpose	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a )

ggml_type_name()

GGML_API const char * ggml_type_name

(

enum ggml_type

type

)

ggml_type_size()

GGML_API size_t ggml_type_size

(

enum ggml_type

type

)

ggml_type_sizef()

GGML_API float ggml_type_sizef

(

enum ggml_type

type

)

ggml_used_mem()

GGML_API size_t ggml_used_mem

(

const struct ggml_context *

ctx

)

ggml_view_1d()

GGML_API struct ggml_tensor * ggml_view_1d	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		int64_t	ne0,
		size_t	offset )

ggml_view_2d()

GGML_API struct ggml_tensor * ggml_view_2d	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		int64_t	ne0,
		int64_t	ne1,
		size_t	nb1,
		size_t	offset )

ggml_view_3d()

GGML_API struct ggml_tensor * ggml_view_3d	(	struct ggml_context *	ctx,
		struct ggml_tensor *	a,
		int64_t	ne0,
		int64_t	ne1,
		int64_t	ne2,
		size_t	nb1,
		size_t	nb2,
		size_t	offset )

ggml_view_tensor()

GGML_API struct ggml_tensor * ggml_view_tensor	(	struct ggml_context *	ctx,
		const struct ggml_tensor *	src )

Variable Documentation

GGML_OBJECT_SIZE

const size_t GGML_OBJECT_SIZE = sizeof(struct ggml_object)

static