llama.cpp/llama_cpp/llama_cpp.py

1869 lines
60 KiB
Python
Raw Normal View History

import sys
import os
2023-03-23 09:33:06 +00:00
import ctypes
2023-04-11 15:59:03 +00:00
from ctypes import (
2023-09-29 02:42:03 +00:00
c_bool,
2023-04-11 15:59:03 +00:00
c_char_p,
2023-09-29 02:42:03 +00:00
c_int,
c_int8,
2023-07-05 05:00:14 +00:00
c_int32,
2023-09-29 02:42:03 +00:00
c_uint8,
2023-07-05 05:00:14 +00:00
c_uint32,
2023-09-29 02:42:03 +00:00
c_size_t,
c_float,
c_double,
2023-04-11 15:59:03 +00:00
c_void_p,
POINTER,
2023-05-05 17:54:22 +00:00
_Pointer, # type: ignore
2023-04-11 15:59:03 +00:00
Structure,
Array,
)
2023-03-23 09:33:06 +00:00
import pathlib
2023-06-26 12:50:38 +00:00
from typing import List, Union
2023-03-23 09:33:06 +00:00
2023-04-11 15:59:03 +00:00
2023-03-23 09:33:06 +00:00
# Load the library
def _load_shared_library(lib_base_name: str):
# Construct the paths to the possible shared library names
_base_path = pathlib.Path(os.path.abspath(os.path.dirname(__file__)))
# Searching for the library in the current directory under the name "libllama" (default name
# for llamacpp) and "llama" (default name for this repo)
_lib_paths: List[pathlib.Path] = []
# Determine the file extension based on the platform
if sys.platform.startswith("linux"):
_lib_paths += [
_base_path / f"lib{lib_base_name}.so",
]
elif sys.platform == "darwin":
_lib_paths += [
_base_path / f"lib{lib_base_name}.so",
_base_path / f"lib{lib_base_name}.dylib",
]
elif sys.platform == "win32":
_lib_paths += [
_base_path / f"{lib_base_name}.dll",
]
else:
raise RuntimeError("Unsupported platform")
2023-04-11 15:59:03 +00:00
if "LLAMA_CPP_LIB" in os.environ:
lib_base_name = os.environ["LLAMA_CPP_LIB"]
2023-04-10 15:12:25 +00:00
_lib = pathlib.Path(lib_base_name)
_base_path = _lib.parent.resolve()
_lib_paths = [_lib.resolve()]
cdll_args = dict() # type: ignore
# Add the library directory to the DLL search path on Windows (if needed)
if sys.platform == "win32" and sys.version_info >= (3, 8):
os.add_dll_directory(str(_base_path))
2023-05-17 19:26:38 +00:00
if "CUDA_PATH" in os.environ:
os.add_dll_directory(os.path.join(os.environ["CUDA_PATH"], "bin"))
os.add_dll_directory(os.path.join(os.environ["CUDA_PATH"], "lib"))
cdll_args["winmode"] = ctypes.RTLD_GLOBAL
# Try to load the shared library, handling potential errors
for _lib_path in _lib_paths:
if _lib_path.exists():
try:
return ctypes.CDLL(str(_lib_path), **cdll_args)
except Exception as e:
raise RuntimeError(f"Failed to load shared library '{_lib_path}': {e}")
2023-04-11 15:59:03 +00:00
raise FileNotFoundError(
f"Shared library with base name '{lib_base_name}' not found"
)
# Specify the base name of the shared library to load
_lib_base_name = "llama"
# Load the library
_lib = _load_shared_library(_lib_base_name)
2023-03-23 09:33:06 +00:00
2023-05-21 21:47:21 +00:00
# Misc
c_float_p = POINTER(c_float)
c_uint8_p = POINTER(c_uint8)
c_size_t_p = POINTER(c_size_t)
# llama.h bindings
2023-06-06 20:23:55 +00:00
GGML_USE_CUBLAS = hasattr(_lib, "ggml_init_cublas")
GGML_CUDA_MAX_DEVICES = 16
LLAMA_MAX_DEVICES = GGML_CUDA_MAX_DEVICES if GGML_USE_CUBLAS else 1
2023-06-06 20:23:55 +00:00
2023-08-24 04:17:00 +00:00
# define LLAMA_DEFAULT_SEED 0xFFFFFFFF
LLAMA_DEFAULT_SEED = 0xFFFFFFFF
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
# define LLAMA_MAX_RNG_STATE (64*1024)
LLAMA_MAX_RNG_STATE = 64 * 1024
2023-08-24 04:17:00 +00:00
# define LLAMA_FILE_MAGIC_GGSN 0x6767736eu // 'ggsn'
LLAMA_FILE_MAGIC_GGSN = 0x6767736E
2023-05-21 21:47:21 +00:00
2023-08-24 04:17:00 +00:00
# define LLAMA_SESSION_MAGIC LLAMA_FILE_MAGIC_GGSN
2023-05-21 21:47:21 +00:00
LLAMA_SESSION_MAGIC = LLAMA_FILE_MAGIC_GGSN
2023-08-24 04:17:00 +00:00
# define LLAMA_SESSION_VERSION 1
LLAMA_SESSION_VERSION = 1
2023-05-01 14:44:28 +00:00
2023-07-05 05:00:14 +00:00
2023-06-26 12:50:38 +00:00
# struct llama_model;
llama_model_p = c_void_p
2023-05-21 21:47:21 +00:00
# struct llama_context;
llama_context_p = c_void_p
2023-09-29 02:42:03 +00:00
# typedef int32_t llama_pos;
llama_pos = c_int32
# typedef int32_t llama_token;
llama_token = c_int32
2023-03-23 09:33:06 +00:00
llama_token_p = POINTER(llama_token)
2023-09-29 02:42:03 +00:00
# typedef int32_t llama_seq_id;
llama_seq_id = c_int32
2023-03-23 09:33:06 +00:00
2023-08-24 04:17:00 +00:00
# enum llama_vocab_type {
# LLAMA_VOCAB_TYPE_SPM = 0, // SentencePiece
# LLAMA_VOCAB_TYPE_BPE = 1, // Byte Pair Encoding
# };
LLAMA_VOCAB_TYPE_SPM = 0
LLAMA_VOCAB_TYPE_BPE = 1
2023-08-24 04:17:00 +00:00
# enum llama_token_type {
# LLAMA_TOKEN_TYPE_UNDEFINED = 0,
# LLAMA_TOKEN_TYPE_NORMAL = 1,
# LLAMA_TOKEN_TYPE_UNKNOWN = 2,
# LLAMA_TOKEN_TYPE_CONTROL = 3,
# LLAMA_TOKEN_TYPE_USER_DEFINED = 4,
# LLAMA_TOKEN_TYPE_UNUSED = 5,
# LLAMA_TOKEN_TYPE_BYTE = 6,
# };
LLAMA_TOKEN_TYPE_UNDEFINED = 0
LLAMA_TOKEN_TYPE_NORMAL = 1
LLAMA_TOKEN_TYPE_UNKNOWN = 2
LLAMA_TOKEN_TYPE_CONTROL = 3
LLAMA_TOKEN_TYPE_USER_DEFINED = 4
LLAMA_TOKEN_TYPE_UNUSED = 5
LLAMA_TOKEN_TYPE_BYTE = 6
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
# // model file types
2023-08-24 04:17:00 +00:00
# enum llama_ftype {
# LLAMA_FTYPE_ALL_F32 = 0,
2023-09-29 02:42:03 +00:00
# LLAMA_FTYPE_MOSTLY_F16 = 1, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q4_0 = 2, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q4_1 = 3, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q4_1_SOME_F16 = 4, // tok_embeddings.weight and output.weight are F16
# // LLAMA_FTYPE_MOSTLY_Q4_2 = 5, // support has been removed
# // LLAMA_FTYPE_MOSTLY_Q4_3 = 6, // support has been removed
# LLAMA_FTYPE_MOSTLY_Q8_0 = 7, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q5_0 = 8, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q5_1 = 9, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q2_K = 10, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q3_K_S = 11, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q3_K_M = 12, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q3_K_L = 13, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q4_K_S = 14, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q4_K_M = 15, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q5_K_S = 16, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q5_K_M = 17, // except 1d tensors
# LLAMA_FTYPE_MOSTLY_Q6_K = 18, // except 1d tensors
2023-08-24 04:17:00 +00:00
# LLAMA_FTYPE_GUESSED = 1024, // not specified in the model file
# };
LLAMA_FTYPE_ALL_F32 = 0
LLAMA_FTYPE_MOSTLY_F16 = 1
LLAMA_FTYPE_MOSTLY_Q4_0 = 2
LLAMA_FTYPE_MOSTLY_Q4_1 = 3
LLAMA_FTYPE_MOSTLY_Q4_1_SOME_F16 = 4
LLAMA_FTYPE_MOSTLY_Q8_0 = 7
LLAMA_FTYPE_MOSTLY_Q5_0 = 8
LLAMA_FTYPE_MOSTLY_Q5_1 = 9
LLAMA_FTYPE_MOSTLY_Q2_K = 10
LLAMA_FTYPE_MOSTLY_Q3_K_S = 11
LLAMA_FTYPE_MOSTLY_Q3_K_M = 12
LLAMA_FTYPE_MOSTLY_Q3_K_L = 13
LLAMA_FTYPE_MOSTLY_Q4_K_S = 14
LLAMA_FTYPE_MOSTLY_Q4_K_M = 15
LLAMA_FTYPE_MOSTLY_Q5_K_S = 16
LLAMA_FTYPE_MOSTLY_Q5_K_M = 17
LLAMA_FTYPE_MOSTLY_Q6_K = 18
LLAMA_FTYPE_GUESSED = 1024
2023-08-24 04:17:00 +00:00
2023-03-24 18:35:41 +00:00
2023-05-21 21:47:21 +00:00
# typedef struct llama_token_data {
# llama_token id; // token id
# float logit; // log-odds of the token
# float p; // probability of the token
# } llama_token_data;
2023-03-23 09:33:06 +00:00
class llama_token_data(Structure):
_fields_ = [
2023-05-21 21:47:21 +00:00
("id", llama_token),
("logit", c_float),
("p", c_float),
2023-03-23 09:33:06 +00:00
]
2023-03-24 18:35:41 +00:00
2023-03-23 09:33:06 +00:00
llama_token_data_p = POINTER(llama_token_data)
2023-05-01 14:44:28 +00:00
2023-05-21 21:47:21 +00:00
# typedef struct llama_token_data_array {
# llama_token_data * data;
# size_t size;
# bool sorted;
# } llama_token_data_array;
2023-05-01 14:44:28 +00:00
class llama_token_data_array(Structure):
_fields_ = [
("data", llama_token_data_p),
("size", c_size_t),
("sorted", c_bool),
]
llama_token_data_array_p = POINTER(llama_token_data_array)
2023-05-21 21:47:21 +00:00
# typedef void (*llama_progress_callback)(float progress, void *ctx);
2023-03-29 01:10:23 +00:00
llama_progress_callback = ctypes.CFUNCTYPE(None, c_float, c_void_p)
2023-03-24 18:35:41 +00:00
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
# // Input data for llama_decode
# // A llama_batch object can contain input about one or many sequences
# // The provided arrays (i.e. token, embd, pos, etc.) must have size of n_tokens
# //
# // - token : the token ids of the input (used when embd is NULL)
# // - embd : token embeddings (i.e. float vector of size n_embd) (used when token is NULL)
# // - pos : the positions of the respective token in the sequence
# // - seq_id : the sequence to which the respective token belongs
# // - logits : if zero, the logits for the respective token will not be output
# //
# typedef struct llama_batch {
# int32_t n_tokens;
# llama_token * token;
# float * embd;
# llama_pos * pos;
# llama_seq_id * seq_id;
# int8_t * logits;
# // NOTE: helpers for smooth API transition - can be deprecated in the future
# // for future-proof code, use the above fields instead and ignore everything below
# //
# // pos[i] = all_pos_0 + i*all_pos_1
# //
# llama_pos all_pos_0; // used if pos == NULL
# llama_pos all_pos_1; // used if pos == NULL
# llama_seq_id all_seq_id; // used if seq_id == NULL
# } llama_batch;
class llama_batch(Structure):
_fields_ = [
("n_tokens", c_int32),
("token", POINTER(llama_token)),
("embd", c_float_p),
("pos", POINTER(llama_pos)),
("seq_id", POINTER(llama_seq_id)),
("logits", POINTER(c_int8)),
("all_pos_0", llama_pos),
("all_pos_1", llama_pos),
("all_seq_id", llama_seq_id),
]
2023-07-15 19:11:01 +00:00
2023-09-29 02:42:03 +00:00
# struct llama_model_params {
# int32_t n_gpu_layers; // number of layers to store in VRAM
# int32_t main_gpu; // the GPU that is used for scratch and small tensors
# const float * tensor_split; // how to split layers across multiple GPUs (size: LLAMA_MAX_DEVICES)
2023-07-15 19:11:01 +00:00
2023-06-20 15:25:10 +00:00
# // called with a progress value between 0 and 1, pass NULL to disable
# llama_progress_callback progress_callback;
# // context pointer passed to the progress callback
# void * progress_callback_user_data;
2023-05-21 21:47:21 +00:00
2023-07-20 22:51:53 +00:00
2023-06-20 15:25:10 +00:00
# // Keep the booleans together to avoid misalignment during copy-by-value.
2023-05-21 21:47:21 +00:00
# bool vocab_only; // only load the vocabulary, no weights
# bool use_mmap; // use mmap if possible
# bool use_mlock; // force system to keep model in RAM
# };
2023-09-29 02:42:03 +00:00
class llama_model_params(Structure):
2023-03-23 09:33:06 +00:00
_fields_ = [
2023-07-05 05:00:14 +00:00
("n_gpu_layers", c_int32),
("main_gpu", c_int32),
2023-08-24 04:17:00 +00:00
("tensor_split", c_float_p),
2023-06-20 15:25:10 +00:00
("progress_callback", llama_progress_callback),
("progress_callback_user_data", c_void_p),
2023-05-21 21:47:21 +00:00
("vocab_only", c_bool),
("use_mmap", c_bool),
("use_mlock", c_bool),
2023-03-23 09:33:06 +00:00
]
2023-03-24 18:35:41 +00:00
2023-09-29 02:42:03 +00:00
# struct llama_context_params {
# uint32_t seed; // RNG seed, -1 for random
2023-09-29 23:58:21 +00:00
# uint32_t n_ctx; // text context, 0 = from model
# uint32_t n_batch; // prompt processing maximum batch size
2023-09-29 02:42:03 +00:00
# uint32_t n_threads; // number of threads to use for generation
# uint32_t n_threads_batch; // number of threads to use for batch processing
# // ref: https://github.com/ggerganov/llama.cpp/pull/2054
2023-09-29 23:58:21 +00:00
# float rope_freq_base; // RoPE base frequency, 0 = from model
# float rope_freq_scale; // RoPE frequency scaling factor, 0 = from model
2023-09-29 02:42:03 +00:00
# // Keep the booleans together to avoid misalignment during copy-by-value.
# bool mul_mat_q; // if true, use experimental mul_mat_q kernels
2023-09-29 23:58:21 +00:00
# bool f16_kv; // use fp16 for KV cache, fp32 otherwise
2023-09-29 02:42:03 +00:00
# bool logits_all; // the llama_eval() call computes all logits, not just the last one
# bool embedding; // embedding mode only
# };
class llama_context_params(Structure):
_fields_ = [
("seed", c_uint32),
("n_ctx", c_uint32),
("n_batch", c_uint32),
("n_threads", c_uint32),
("n_threads_batch", c_uint32),
("rope_freq_base", c_float),
("rope_freq_scale", c_float),
("mul_mat_q", c_bool),
("f16_kv", c_bool),
("logits_all", c_bool),
("embedding", c_bool),
]
2023-03-23 09:33:06 +00:00
2023-08-24 04:17:00 +00:00
# // Signature for logging events
# // Note that text includes the new line character at the end for most events.
# // If your logging mechanism cannot handle that, check if the last character is '\n' and strip it
# // if it exists.
# // It might not exist for progress report where '.' is output repeatedly.
# typedef void (*llama_log_callback)(enum llama_log_level level, const char * text, void * user_data);
llama_log_callback = ctypes.CFUNCTYPE(None, c_int, c_char_p, c_void_p)
2023-03-23 09:33:06 +00:00
2023-03-24 18:35:41 +00:00
2023-06-10 16:17:38 +00:00
# // model quantization parameters
# typedef struct llama_model_quantize_params {
# int nthread; // number of threads to use for quantizing, if <=0 will use std::thread::hardware_concurrency()
2023-08-24 04:17:00 +00:00
# enum llama_ftype ftype; // quantize to this llama_ftype
2023-06-10 16:17:38 +00:00
# bool allow_requantize; // allow quantizing non-f32/f16 tensors
# bool quantize_output_tensor; // quantize output.weight
2023-09-01 18:26:13 +00:00
# bool only_copy; // only copy tensors - ftype, allow_requantize and quantize_output_tensor are ignored
2023-06-10 16:17:38 +00:00
# } llama_model_quantize_params;
class llama_model_quantize_params(Structure):
_fields_ = [
("nthread", c_int),
("ftype", c_int),
("allow_requantize", c_bool),
("quantize_output_tensor", c_bool),
2023-09-01 18:26:13 +00:00
("only_copy", c_bool),
2023-06-10 16:17:38 +00:00
]
2023-07-24 17:04:34 +00:00
# // grammar types
# struct llama_grammar;
llama_grammar_p = c_void_p
# // grammar element type
# enum llama_gretype {
# // end of rule definition
# LLAMA_GRETYPE_END = 0,
# // start of alternate definition for rule
# LLAMA_GRETYPE_ALT = 1,
# // non-terminal element: reference to rule
# LLAMA_GRETYPE_RULE_REF = 2,
# // terminal element: character (code point)
# LLAMA_GRETYPE_CHAR = 3,
# // inverse char(s) ([^a], [^a-b] [^abc])
# LLAMA_GRETYPE_CHAR_NOT = 4,
# // modifies a preceding LLAMA_GRETYPE_CHAR or LLAMA_GRETYPE_CHAR_ALT to
# // be an inclusive range ([a-z])
# LLAMA_GRETYPE_CHAR_RNG_UPPER = 5,
# // modifies a preceding LLAMA_GRETYPE_CHAR or
# // LLAMA_GRETYPE_CHAR_RNG_UPPER to add an alternate char to match ([ab], [a-zA])
# LLAMA_GRETYPE_CHAR_ALT = 6,
# };
LLAMA_GRETYPE_END = 0
LLAMA_GRETYPE_ALT = 1
LLAMA_GRETYPE_RULE_REF = 2
LLAMA_GRETYPE_CHAR = 3
LLAMA_GRETYPE_CHAR_NOT = 4
LLAMA_GRETYPE_CHAR_RNG_UPPER = 5
LLAMA_GRETYPE_CHAR_ALT = 6
2023-07-24 17:04:34 +00:00
# typedef struct llama_grammar_element {
# enum llama_gretype type;
# uint32_t value; // Unicode code point or rule ID
# } llama_grammar_element;
class llama_grammar_element(Structure):
_fields_ = [
("type", c_int),
("value", c_uint32),
]
llama_grammar_element_p = POINTER(llama_grammar_element)
2023-07-06 21:57:56 +00:00
# // performance timing information
# struct llama_timings {
# double t_start_ms;
# double t_end_ms;
# double t_load_ms;
# double t_sample_ms;
# double t_p_eval_ms;
# double t_eval_ms;
# int32_t n_sample;
# int32_t n_p_eval;
# int32_t n_eval;
# };
class llama_timings(Structure):
_fields_ = [
("t_start_ms", c_double),
("t_end_ms", c_double),
("t_load_ms", c_double),
("t_sample_ms", c_double),
("t_p_eval_ms", c_double),
("t_eval_ms", c_double),
("n_sample", c_int32),
("n_p_eval", c_int32),
("n_eval", c_int32),
]
2023-09-29 02:42:03 +00:00
# // Helpers for getting default parameters
# LLAMA_API struct llama_model_params llama_model_default_params(void);
def llama_model_default_params() -> llama_model_params:
return _lib.llama_model_default_params()
_lib.llama_model_default_params.argtypes = []
_lib.llama_model_default_params.restype = llama_model_params
2023-08-24 04:17:00 +00:00
# LLAMA_API struct llama_context_params llama_context_default_params(void);
2023-03-23 09:33:06 +00:00
def llama_context_default_params() -> llama_context_params:
2023-03-29 01:10:23 +00:00
return _lib.llama_context_default_params()
2023-03-23 09:33:06 +00:00
2023-03-24 18:35:41 +00:00
_lib.llama_context_default_params.argtypes = []
_lib.llama_context_default_params.restype = llama_context_params
2023-04-11 15:59:03 +00:00
2023-08-24 04:17:00 +00:00
# LLAMA_API struct llama_model_quantize_params llama_model_quantize_default_params(void);
2023-06-10 16:17:38 +00:00
def llama_model_quantize_default_params() -> llama_model_quantize_params:
return _lib.llama_model_quantize_default_params()
_lib.llama_model_quantize_default_params.argtypes = []
_lib.llama_model_quantize_default_params.restype = llama_model_quantize_params
2023-05-21 21:47:21 +00:00
# // Initialize the llama + ggml backend
2023-06-29 05:08:15 +00:00
# // If numa is true, use NUMA optimizations
2023-05-21 21:47:21 +00:00
# // Call once at the start of the program
2023-07-15 19:11:01 +00:00
# LLAMA_API void llama_backend_init(bool numa);
def llama_backend_init(numa: Union[c_bool, bool]):
2023-07-15 19:11:01 +00:00
return _lib.llama_backend_init(numa)
_lib.llama_backend_init.argtypes = [c_bool]
_lib.llama_backend_init.restype = None
2023-05-21 21:47:21 +00:00
2023-07-15 19:11:01 +00:00
# // Call once at the end of the program - currently only used for MPI
2023-08-24 04:17:00 +00:00
# LLAMA_API void llama_backend_free(void);
2023-07-15 19:11:01 +00:00
def llama_backend_free():
return _lib.llama_backend_free()
2023-05-21 21:47:21 +00:00
2023-07-15 19:11:01 +00:00
_lib.llama_backend_free.argtypes = []
_lib.llama_backend_free.restype = None
2023-05-21 21:47:21 +00:00
2023-06-26 12:50:38 +00:00
# LLAMA_API struct llama_model * llama_load_model_from_file(
2023-08-24 04:17:00 +00:00
# const char * path_model,
2023-06-26 12:50:38 +00:00
# struct llama_context_params params);
def llama_load_model_from_file(
2023-09-29 02:42:03 +00:00
path_model: bytes, params: llama_model_params
2023-06-26 12:50:38 +00:00
) -> llama_model_p:
return _lib.llama_load_model_from_file(path_model, params)
2023-06-26 12:50:38 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_load_model_from_file.argtypes = [c_char_p, llama_model_params]
2023-06-26 12:50:38 +00:00
_lib.llama_load_model_from_file.restype = llama_model_p
# LLAMA_API void llama_free_model(struct llama_model * model);
def llama_free_model(model: llama_model_p):
return _lib.llama_free_model(model)
_lib.llama_free_model.argtypes = [llama_model_p]
_lib.llama_free_model.restype = None
# LLAMA_API struct llama_context * llama_new_context_with_model(
2023-08-24 04:17:00 +00:00
# struct llama_model * model,
2023-06-26 12:50:38 +00:00
# struct llama_context_params params);
def llama_new_context_with_model(
model: llama_model_p, params: llama_context_params
) -> llama_context_p:
return _lib.llama_new_context_with_model(model, params)
_lib.llama_new_context_with_model.argtypes = [llama_model_p, llama_context_params]
_lib.llama_new_context_with_model.restype = llama_context_p
2023-08-24 04:17:00 +00:00
# // Frees all allocated memory
# LLAMA_API void llama_free(struct llama_context * ctx);
def llama_free(ctx: llama_context_p):
return _lib.llama_free(ctx)
_lib.llama_free.argtypes = [llama_context_p]
_lib.llama_free.restype = None
# LLAMA_API int64_t llama_time_us(void);
2023-05-21 21:47:21 +00:00
def llama_time_us() -> int:
return _lib.llama_time_us()
_lib.llama_time_us.argtypes = []
_lib.llama_time_us.restype = ctypes.c_int64
2023-08-24 04:17:00 +00:00
# LLAMA_API int llama_max_devices (void);
def llama_max_devices() -> int:
return _lib.llama_max_devices()
2023-03-23 09:33:06 +00:00
2023-03-24 18:35:41 +00:00
2023-08-24 04:17:00 +00:00
_lib.llama_max_devices.argtypes = []
_lib.llama_max_devices.restype = c_int
2023-03-24 18:59:29 +00:00
2023-08-24 04:17:00 +00:00
# LLAMA_API bool llama_mmap_supported (void);
def llama_mmap_supported() -> bool:
return _lib.llama_mmap_supported()
2023-03-23 09:33:06 +00:00
2023-03-24 18:35:41 +00:00
2023-08-24 04:17:00 +00:00
_lib.llama_mmap_supported.argtypes = []
_lib.llama_mmap_supported.restype = c_bool
# LLAMA_API bool llama_mlock_supported(void);
def llama_mlock_supported() -> bool:
return _lib.llama_mlock_supported()
_lib.llama_mlock_supported.argtypes = []
_lib.llama_mlock_supported.restype = c_bool
2023-09-29 02:42:03 +00:00
# LLAMA_API const struct llama_model * llama_get_model(const struct llama_context * ctx);
def llama_get_model(ctx: llama_context_p) -> llama_model_p:
return _lib.llama_get_model(ctx)
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_get_model.argtypes = [llama_context_p]
_lib.llama_get_model.restype = llama_model_p
2023-08-24 04:17:00 +00:00
2023-09-09 16:12:32 +00:00
# LLAMA_API int llama_n_ctx (const struct llama_context * ctx);
2023-08-24 04:17:00 +00:00
def llama_n_ctx(ctx: llama_context_p) -> int:
return _lib.llama_n_ctx(ctx)
_lib.llama_n_ctx.argtypes = [llama_context_p]
_lib.llama_n_ctx.restype = c_int
2023-09-29 02:42:03 +00:00
# LLAMA_API enum llama_vocab_type llama_vocab_type(const struct llama_model * model);
def llama_vocab_type(model: llama_model_p) -> int:
return _lib.llama_vocab_type(model)
2023-09-09 16:12:32 +00:00
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_vocab_type.argtypes = [llama_model_p]
2023-08-24 22:01:42 +00:00
_lib.llama_vocab_type.restype = c_int
2023-09-29 02:42:03 +00:00
# LLAMA_API int llama_n_vocab (const struct llama_model * model);
def llama_n_vocab(model: llama_model_p) -> int:
return _lib.llama_n_vocab(model)
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_n_vocab.argtypes = [llama_model_p]
_lib.llama_n_vocab.restype = c_int
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
# LLAMA_API int llama_n_ctx_train(const struct llama_model * model);
def llama_n_ctx_train(model: llama_model_p) -> int:
return _lib.llama_n_ctx_train(model)
2023-09-09 16:12:32 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_n_ctx_train.argtypes = [llama_model_p]
_lib.llama_n_ctx_train.restype = c_int
2023-09-09 16:12:32 +00:00
2023-09-29 02:42:03 +00:00
# LLAMA_API int llama_n_embd (const struct llama_model * model);
def llama_n_embd(model: llama_model_p) -> int:
return _lib.llama_n_embd(model)
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_n_embd.argtypes = [llama_model_p]
_lib.llama_n_embd.restype = c_int
2023-08-24 04:17:00 +00:00
# // Get a string describing the model type
2023-08-25 18:35:53 +00:00
# LLAMA_API int llama_model_desc(const struct llama_model * model, char * buf, size_t buf_size);
2023-09-29 02:42:03 +00:00
def llama_model_desc(
model: llama_model_p, buf: bytes, buf_size: Union[c_size_t, int]
) -> int:
2023-08-25 18:35:53 +00:00
return _lib.llama_model_desc(model, buf, buf_size)
2023-08-24 04:17:00 +00:00
2023-08-25 18:35:53 +00:00
_lib.llama_model_desc.argtypes = [llama_model_p, c_char_p, c_size_t]
_lib.llama_model_desc.restype = c_int
# // Returns the total size of all the tensors in the model in bytes
# LLAMA_API uint64_t llama_model_size(const struct llama_model * model);
def llama_model_size(model: llama_model_p) -> int:
return _lib.llama_model_size(model)
2023-08-24 04:17:00 +00:00
2023-08-25 18:35:53 +00:00
_lib.llama_model_size.argtypes = [llama_model_p]
_lib.llama_model_size.restype = ctypes.c_uint64
2023-08-24 04:17:00 +00:00
2023-08-25 18:35:53 +00:00
# // Returns the total number of parameters in the model
# LLAMA_API uint64_t llama_model_n_params(const struct llama_model * model);
def llama_model_n_params(model: llama_model_p) -> int:
return _lib.llama_model_n_params(model)
_lib.llama_model_n_params.argtypes = [llama_model_p]
_lib.llama_model_n_params.restype = ctypes.c_uint64
2023-03-24 18:59:29 +00:00
2023-09-29 02:42:03 +00:00
# // Get a llama model tensor
# LLAMA_API struct ggml_tensor * llama_get_model_tensor(struct llama_model * model, const char * name);
def llama_get_model_tensor(
model: llama_model_p, name: Union[c_char_p, bytes]
) -> c_void_p:
return _lib.llama_get_model_tensor(model, name)
_lib.llama_get_model_tensor.argtypes = [llama_model_p, c_char_p]
_lib.llama_get_model_tensor.restype = c_void_p
2023-06-10 16:17:38 +00:00
# // Returns 0 on success
2023-05-21 21:47:21 +00:00
# LLAMA_API int llama_model_quantize(
# const char * fname_inp,
# const char * fname_out,
2023-06-10 16:17:38 +00:00
# const llama_model_quantize_params * params);
2023-04-22 23:50:28 +00:00
def llama_model_quantize(
2023-06-10 16:17:38 +00:00
fname_inp: bytes,
fname_out: bytes,
params, # type: POINTER(llama_model_quantize_params) # type: ignore
) -> int:
2023-06-10 16:17:38 +00:00
return _lib.llama_model_quantize(fname_inp, fname_out, params)
2023-03-23 09:33:06 +00:00
2023-03-24 18:35:41 +00:00
2023-06-10 16:17:38 +00:00
_lib.llama_model_quantize.argtypes = [
c_char_p,
c_char_p,
POINTER(llama_model_quantize_params),
]
_lib.llama_model_quantize.restype = c_int
2023-04-11 15:59:03 +00:00
2023-09-29 02:42:03 +00:00
# // Apply a LoRA adapter to a loaded model
# // path_base_model is the path to a higher quality model to use as a base for
# // the layers modified by the adapter. Can be NULL to use the current loaded model.
# // The model needs to be reloaded before applying a new adapter, otherwise the adapter
# // will be applied on top of the previous one
# // Returns 0 on success
# LLAMA_API DEPRECATED(int llama_apply_lora_from_file(
2023-05-21 21:47:21 +00:00
# struct llama_context * ctx,
# const char * path_lora,
2023-09-29 02:42:03 +00:00
# float scale,
2023-05-21 21:47:21 +00:00
# const char * path_base_model,
2023-09-29 02:42:03 +00:00
# int n_threads),
# "use llama_model_apply_lora_from_file instead");
def llama_apply_lora_from_file(
2023-04-22 23:50:28 +00:00
ctx: llama_context_p,
path_lora: Union[c_char_p, bytes],
2023-09-29 02:42:03 +00:00
scale: Union[c_float, float],
path_base_model: Union[c_char_p, bytes],
n_threads: Union[c_int, int],
) -> int:
2023-09-29 02:42:03 +00:00
return _lib.llama_apply_lora_from_file(
ctx, path_lora, scale, path_base_model, n_threads
)
2023-09-29 02:42:03 +00:00
_lib.llama_apply_lora_from_file.argtypes = [
llama_context_p,
c_char_p,
c_float,
c_char_p,
c_int,
]
_lib.llama_apply_lora_from_file.restype = c_int
2023-06-26 12:50:38 +00:00
# LLAMA_API int llama_model_apply_lora_from_file(
# const struct llama_model * model,
2023-09-29 02:42:03 +00:00
# const char * path_lora,
# float scale,
# const char * path_base_model,
# int n_threads);
2023-06-26 12:50:38 +00:00
def llama_model_apply_lora_from_file(
model: llama_model_p,
path_lora: Union[c_char_p, bytes],
2023-09-29 02:42:03 +00:00
scale: Union[c_float, float],
2023-06-26 12:50:38 +00:00
path_base_model: Union[c_char_p, bytes],
n_threads: Union[c_int, int],
2023-06-26 12:50:38 +00:00
) -> int:
return _lib.llama_model_apply_lora_from_file(
2023-09-29 02:42:03 +00:00
model, path_lora, scale, path_base_model, n_threads
2023-06-26 12:50:38 +00:00
)
_lib.llama_model_apply_lora_from_file.argtypes = [
llama_model_p,
c_char_p,
2023-09-29 02:42:03 +00:00
c_float,
2023-06-26 12:50:38 +00:00
c_char_p,
c_int,
]
_lib.llama_model_apply_lora_from_file.restype = c_int
2023-09-29 02:42:03 +00:00
# //
# // KV cache
# //
2023-09-29 02:42:03 +00:00
# // Returns the number of tokens in the KV cache
# LLAMA_API DEPRECATED(int llama_get_kv_cache_token_count(const struct llama_context * ctx),
# "avoid using this, it will be removed in the future, instead - count the tokens in user code");
def llama_get_kv_cache_token_count(ctx: llama_context_p) -> int:
return _lib.llama_get_kv_cache_token_count(ctx)
2023-04-11 15:59:03 +00:00
_lib.llama_get_kv_cache_token_count.argtypes = [llama_context_p]
_lib.llama_get_kv_cache_token_count.restype = c_int
2023-04-28 19:32:43 +00:00
2023-09-29 02:42:03 +00:00
# // Remove all tokens data of cells in [c0, c1)
# LLAMA_API void llama_kv_cache_tokens_rm(
# struct llama_context * ctx,
# int32_t c0,
# int32_t c1);
def llama_kv_cache_tokens_rm(
ctx: llama_context_p, c0: Union[c_int32, int], c1: Union[c_int32, int]
):
return _lib.llama_kv_cache_tokens_rm(ctx, c0, c1)
2023-04-27 00:00:54 +00:00
2023-04-28 19:32:43 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_kv_cache_tokens_rm.argtypes = [llama_context_p, c_int32, c_int32]
_lib.llama_kv_cache_tokens_rm.restype = None
# // Removes all tokens that belong to the specified sequence and have positions in [p0, p1)
# LLAMA_API void llama_kv_cache_seq_rm(
# struct llama_context * ctx,
# llama_seq_id seq_id,
# llama_pos p0,
# llama_pos p1);
def llama_kv_cache_seq_rm(
ctx: llama_context_p,
seq_id: llama_seq_id,
p0: Union[llama_pos, int],
p1: Union[llama_pos, int],
):
return _lib.llama_kv_cache_seq_rm(ctx, seq_id, p0, p1)
_lib.llama_kv_cache_seq_rm.argtypes = [
llama_context_p,
llama_seq_id,
llama_pos,
llama_pos,
]
_lib.llama_kv_cache_seq_rm.restype = None
# // Copy all tokens that belong to the specified sequence to another sequence
# // Note that this does not allocate extra KV cache memory - it simply assigns the tokens to the new sequence
# LLAMA_API void llama_kv_cache_seq_cp(
# struct llama_context * ctx,
# llama_seq_id seq_id_src,
# llama_seq_id seq_id_dst,
# llama_pos p0,
# llama_pos p1);
def llama_kv_cache_seq_cp(
ctx: llama_context_p,
seq_id_src: llama_seq_id,
seq_id_dst: llama_seq_id,
p0: Union[llama_pos, int],
p1: Union[llama_pos, int],
):
return _lib.llama_kv_cache_seq_cp(ctx, seq_id_src, seq_id_dst, p0, p1)
_lib.llama_kv_cache_seq_cp.argtypes = [
llama_context_p,
llama_seq_id,
llama_seq_id,
llama_pos,
llama_pos,
]
_lib.llama_kv_cache_seq_cp.restype = None
# // Removes all tokens that do not belong to the specified sequence
# LLAMA_API void llama_kv_cache_seq_keep(
# struct llama_context * ctx,
# llama_seq_id seq_id);
def llama_kv_cache_seq_keep(
ctx: llama_context_p,
seq_id: llama_seq_id,
):
return _lib.llama_kv_cache_seq_keep(ctx, seq_id)
_lib.llama_kv_cache_seq_keep.argtypes = [llama_context_p, llama_seq_id]
_lib.llama_kv_cache_seq_keep.restype = None
# // Adds relative position "delta" to all tokens that belong to the specified sequence and have positions in [p0, p1)
# // If the KV cache is RoPEd, the KV data is updated accordingly
# LLAMA_API void llama_kv_cache_seq_shift(
# struct llama_context * ctx,
# llama_seq_id seq_id,
# llama_pos p0,
# llama_pos p1,
# llama_pos delta);
def llama_kv_cache_seq_shift(
ctx: llama_context_p,
seq_id: llama_seq_id,
p0: Union[llama_pos, int],
p1: Union[llama_pos, int],
delta: Union[llama_pos, int],
):
return _lib.llama_kv_cache_seq_shift(ctx, seq_id, p0, p1, delta)
_lib.llama_kv_cache_seq_shift.argtypes = [
llama_context_p,
llama_seq_id,
llama_pos,
llama_pos,
llama_pos,
]
_lib.llama_kv_cache_seq_shift.restype = None
# //
# // State / sessions
# //
2023-04-28 19:32:43 +00:00
2023-05-03 13:33:30 +00:00
# Returns the maximum size in bytes of the state (rng, logits, embedding
# and kv_cache) - will often be smaller after compacting tokens
2023-05-21 21:47:21 +00:00
# LLAMA_API size_t llama_get_state_size(const struct llama_context * ctx);
def llama_get_state_size(ctx: llama_context_p) -> int:
2023-04-22 23:50:28 +00:00
return _lib.llama_get_state_size(ctx)
_lib.llama_get_state_size.argtypes = [llama_context_p]
_lib.llama_get_state_size.restype = c_size_t
# Copies the state to the specified destination address.
# Destination needs to have allocated enough memory.
# Returns the number of bytes copied
2023-09-29 02:42:03 +00:00
# LLAMA_API size_t llama_copy_state_data(
# struct llama_context * ctx,
# uint8_t * dst);
2023-05-05 18:12:26 +00:00
def llama_copy_state_data(
2023-05-14 04:04:22 +00:00
ctx: llama_context_p, dst # type: Array[c_uint8]
2023-05-07 23:30:14 +00:00
) -> int:
2023-05-14 04:04:22 +00:00
return _lib.llama_copy_state_data(ctx, dst)
2023-04-22 23:50:28 +00:00
2023-05-05 18:04:12 +00:00
_lib.llama_copy_state_data.argtypes = [llama_context_p, c_uint8_p]
2023-04-22 23:50:28 +00:00
_lib.llama_copy_state_data.restype = c_size_t
# Set the state reading from the specified address
# Returns the number of bytes read
2023-09-29 02:42:03 +00:00
# LLAMA_API size_t llama_set_state_data(
# struct llama_context * ctx,
# uint8_t * src);
def llama_set_state_data(
ctx: llama_context_p, src # type: Array[c_uint8]
2023-05-07 23:30:14 +00:00
) -> int:
2023-04-22 23:50:28 +00:00
return _lib.llama_set_state_data(ctx, src)
2023-05-05 18:04:12 +00:00
_lib.llama_set_state_data.argtypes = [llama_context_p, c_uint8_p]
2023-04-22 23:50:28 +00:00
_lib.llama_set_state_data.restype = c_size_t
2023-04-28 19:32:43 +00:00
# Save/load session file
2023-09-29 02:42:03 +00:00
# LLAMA_API bool llama_load_session_file(
# struct llama_context * ctx,
# const char * path_session,
# llama_token * tokens_out,
# size_t n_token_capacity,
# size_t * n_token_count_out);
2023-04-28 19:32:43 +00:00
def llama_load_session_file(
ctx: llama_context_p,
path_session: bytes,
2023-05-05 18:12:26 +00:00
tokens_out, # type: Array[llama_token]
n_token_capacity: Union[c_size_t, int],
2023-05-05 18:12:26 +00:00
n_token_count_out, # type: _Pointer[c_size_t]
) -> int:
2023-04-28 19:32:43 +00:00
return _lib.llama_load_session_file(
ctx, path_session, tokens_out, n_token_capacity, n_token_count_out
)
_lib.llama_load_session_file.argtypes = [
llama_context_p,
c_char_p,
llama_token_p,
c_size_t,
2023-05-05 18:04:12 +00:00
c_size_t_p,
2023-04-28 19:32:43 +00:00
]
_lib.llama_load_session_file.restype = c_size_t
2023-09-29 02:42:03 +00:00
# LLAMA_API bool llama_save_session_file(
# struct llama_context * ctx,
# const char * path_session,
# const llama_token * tokens,
# size_t n_token_count);
2023-04-28 19:32:43 +00:00
def llama_save_session_file(
ctx: llama_context_p,
path_session: bytes,
2023-05-05 18:12:26 +00:00
tokens, # type: Array[llama_token]
n_token_count: Union[c_size_t, int],
) -> int:
2023-04-28 19:32:43 +00:00
return _lib.llama_save_session_file(ctx, path_session, tokens, n_token_count)
_lib.llama_save_session_file.argtypes = [
llama_context_p,
c_char_p,
llama_token_p,
c_size_t,
]
_lib.llama_save_session_file.restype = c_size_t
2023-09-29 02:42:03 +00:00
# //
# // Decoding
# //
2023-04-28 19:32:43 +00:00
2023-09-29 02:42:03 +00:00
# // Run the llama inference to obtain the logits and probabilities for the next token(s).
# // tokens + n_tokens is the provided batch of new tokens to process
# // n_past is the number of tokens to use from previous eval calls
# // Returns 0 on success
# // DEPRECATED: use llama_decode() instead
# LLAMA_API DEPRECATED(int llama_eval(
2023-05-21 21:47:21 +00:00
# struct llama_context * ctx,
2023-09-29 02:42:03 +00:00
# llama_token * tokens,
# int32_t n_tokens,
# int n_past),
# "use llama_decode() instead");
2023-03-24 18:35:41 +00:00
def llama_eval(
ctx: llama_context_p,
2023-05-05 18:12:26 +00:00
tokens, # type: Array[llama_token]
n_tokens: Union[c_int, int],
n_past: Union[c_int, int],
) -> int:
2023-09-29 02:42:03 +00:00
return _lib.llama_eval(ctx, tokens, n_tokens, n_past)
2023-03-23 09:33:06 +00:00
2023-03-24 18:35:41 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_eval.argtypes = [llama_context_p, llama_token_p, c_int, c_int]
_lib.llama_eval.restype = c_int
2023-06-29 05:08:15 +00:00
# // Same as llama_eval, but use float matrix input directly.
2023-09-29 02:42:03 +00:00
# // DEPRECATED: use llama_decode() instead
# LLAMA_API DEPRECATED(int llama_eval_embd(
2023-06-29 05:08:15 +00:00
# struct llama_context * ctx,
2023-09-29 02:42:03 +00:00
# float * embd,
# int32_t n_tokens,
# int n_past),
# "use llama_decode() instead");
2023-06-29 05:08:15 +00:00
def llama_eval_embd(
ctx: llama_context_p,
embd, # type: Array[c_float]
n_tokens: Union[c_int, int],
n_past: Union[c_int, int],
2023-06-29 05:08:15 +00:00
) -> int:
2023-09-29 02:42:03 +00:00
return _lib.llama_eval_embd(ctx, embd, n_tokens, n_past)
2023-06-29 05:08:15 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_eval_embd.argtypes = [llama_context_p, c_float_p, c_int, c_int]
2023-06-29 05:08:15 +00:00
_lib.llama_eval_embd.restype = c_int
2023-09-29 02:42:03 +00:00
# // Return batch for single sequence of tokens starting at pos_0
# //
# // NOTE: this is a helper function to facilitate transition to the new batch API - avoid using it
# //
# LLAMA_API struct llama_batch llama_batch_get_one(
# llama_token * tokens,
# int32_t n_tokens,
# llama_pos pos_0,
# llama_seq_id seq_id);
def llama_batch_get_one(
tokens, # type: Array[llama_token]
n_tokens: Union[c_int, int],
pos_0: Union[llama_pos, int],
seq_id: llama_seq_id,
) -> llama_batch:
return _lib.llama_batch_get_one(tokens, n_tokens, pos_0, seq_id)
_lib.llama_batch_get_one.argtypes = [
llama_token_p,
c_int,
llama_pos,
llama_seq_id,
]
_lib.llama_batch_get_one.restype = llama_batch
# // Allocates a batch of tokens on the heap
# // The batch has to be freed with llama_batch_free()
# // If embd != 0, llama_batch.embd will be allocated with size of n_tokens * embd * sizeof(float)
# // Otherwise, llama_batch.token will be allocated to store n_tokens llama_token
# // The rest of the llama_batch members are allocated with size n_tokens
# // All members are left uninitialized
# LLAMA_API struct llama_batch llama_batch_init(
# int32_t n_tokens,
# int32_t embd);
def llama_batch_init(
n_tokens: Union[c_int, int], embd: Union[c_int, int]
) -> llama_batch:
return _lib.llama_batch_init(n_tokens, embd)
_lib.llama_batch_init.argtypes = [c_int, c_int]
_lib.llama_batch_init.restype = llama_batch
# // Frees a batch of tokens allocated with llama_batch_init()
# LLAMA_API void llama_batch_free(struct llama_batch batch);
def llama_batch_free(batch: llama_batch):
return _lib.llama_batch_free(batch)
_lib.llama_batch_free.argtypes = [llama_batch]
_lib.llama_batch_free.restype = None
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
# // Positive return values does not mean a fatal error, but rather a warning.
# // 0 - success
# // 1 - could not find a KV slot for the batch (try reducing the size of the batch or increase the context)
# // < 0 - error
# LLAMA_API int llama_decode(
# struct llama_context * ctx,
# struct llama_batch batch);
def llama_decode(ctx: llama_context_p, batch: llama_batch) -> int:
return _lib.llama_decode(ctx, batch)
_lib.llama_decode.argtypes = [llama_context_p, llama_batch]
_lib.llama_decode.restype = c_int
# // Set the number of threads used for decoding
# // n_threads is the number of threads used for generation (single token)
# // n_threads_batch is the number of threads used for prompt and batch processing (multiple tokens)
# LLAMA_API void llama_set_n_threads(struct llama_context * ctx, uint32_t n_threads, uint32_t n_threads_batch);
def llama_set_n_threads(
ctx: llama_context_p,
n_threads: Union[c_uint32, int],
n_threads_batch: Union[c_uint32, int],
):
return _lib.llama_set_n_threads(ctx, n_threads, n_threads_batch)
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_set_n_threads.argtypes = [llama_context_p, c_uint32, c_uint32]
_lib.llama_set_n_threads.restype = None
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
# // Token logits obtained from the last call to llama_eval()
# // The logits for the last token are stored in the last row
# // Logits for which llama_batch.logits[i] == 0 are undefined
# // Rows: n_tokens provided with llama_batch
# // Cols: n_vocab
2023-08-24 04:17:00 +00:00
# LLAMA_API float * llama_get_logits(struct llama_context * ctx);
def llama_get_logits(
2023-03-24 18:35:41 +00:00
ctx: llama_context_p,
2023-08-24 04:17:00 +00:00
): # type: (...) -> Array[float] # type: ignore
return _lib.llama_get_logits(ctx)
2023-03-23 09:33:06 +00:00
2023-03-24 18:35:41 +00:00
2023-08-24 04:17:00 +00:00
_lib.llama_get_logits.argtypes = [llama_context_p]
_lib.llama_get_logits.restype = c_float_p
2023-09-29 02:42:03 +00:00
# // Logits for the ith token. Equivalent to:
# // llama_get_logits(ctx) + i*n_vocab
# LLAMA_API float * llama_get_logits_ith(struct llama_context * ctx, int32_t i);
def llama_get_logits_ith(
ctx: llama_context_p, i: Union[c_int32, int]
): # type: (...) -> Array[float] # type: ignore
return _lib.llama_get_logits_ith(ctx, i)
_lib.llama_get_logits_ith.argtypes = [llama_context_p, c_int32]
_lib.llama_get_logits_ith.restype = c_float_p
2023-08-24 04:17:00 +00:00
# Get the embeddings for the input
# shape: [n_embd] (1-dimensional)
# LLAMA_API float * llama_get_embeddings(struct llama_context * ctx);
def llama_get_embeddings(
ctx: llama_context_p,
): # type: (...) -> Array[float] # type: ignore
return _lib.llama_get_embeddings(ctx)
2023-07-15 19:11:01 +00:00
2023-08-24 04:17:00 +00:00
_lib.llama_get_embeddings.argtypes = [llama_context_p]
_lib.llama_get_embeddings.restype = c_float_p
2023-03-23 09:33:06 +00:00
2023-03-24 18:35:41 +00:00
2023-08-24 04:17:00 +00:00
# //
# // Vocab
# //
2023-08-24 04:17:00 +00:00
# LLAMA_API const char * llama_token_get_text(const struct llama_context * ctx, llama_token token);
def llama_token_get_text(ctx: llama_context_p, token: llama_token) -> bytes:
return _lib.llama_token_get_text(ctx, token)
2023-03-23 09:33:06 +00:00
2023-03-24 18:35:41 +00:00
2023-08-24 04:17:00 +00:00
_lib.llama_token_get_text.argtypes = [llama_context_p, llama_token]
_lib.llama_token_get_text.restype = c_char_p
2023-03-24 18:59:29 +00:00
2023-08-24 04:17:00 +00:00
# LLAMA_API float llama_token_get_score(const struct llama_context * ctx, llama_token token);
def llama_token_get_score(ctx: llama_context_p, token: llama_token) -> float:
return _lib.llama_token_get_score(ctx, token)
2023-03-25 20:26:03 +00:00
2023-08-24 04:17:00 +00:00
_lib.llama_token_get_score.argtypes = [llama_context_p, llama_token]
_lib.llama_token_get_score.restype = c_float
2023-03-25 20:26:03 +00:00
2023-08-24 22:01:42 +00:00
# LLAMA_API enum llama_token_type llama_token_get_type(const struct llama_context * ctx, llama_token token);
2023-08-24 04:17:00 +00:00
def llama_token_get_type(ctx: llama_context_p, token: llama_token) -> int:
return _lib.llama_token_get_type(ctx, token)
2023-07-15 19:11:01 +00:00
2023-08-24 04:17:00 +00:00
_lib.llama_token_get_type.argtypes = [llama_context_p, llama_token]
_lib.llama_token_get_type.restype = ctypes.c_int
2023-07-15 19:11:01 +00:00
2023-08-24 04:17:00 +00:00
# // Special tokens
2023-07-15 19:11:01 +00:00
2023-08-24 04:17:00 +00:00
# LLAMA_API llama_token llama_token_bos(const struct llama_context * ctx); // beginning-of-sentence
2023-09-29 02:42:03 +00:00
def llama_token_bos(ctx: llama_context_p) -> int:
2023-08-24 04:17:00 +00:00
return _lib.llama_token_bos(ctx)
2023-07-15 19:11:01 +00:00
2023-08-24 04:17:00 +00:00
_lib.llama_token_bos.argtypes = [llama_context_p]
_lib.llama_token_bos.restype = llama_token
2023-07-15 19:11:01 +00:00
2023-08-24 04:17:00 +00:00
# LLAMA_API llama_token llama_token_eos(const struct llama_context * ctx); // end-of-sentence
2023-09-29 02:42:03 +00:00
def llama_token_eos(ctx: llama_context_p) -> int:
2023-08-24 04:17:00 +00:00
return _lib.llama_token_eos(ctx)
2023-07-15 19:11:01 +00:00
2023-08-24 04:17:00 +00:00
_lib.llama_token_eos.argtypes = [llama_context_p]
_lib.llama_token_eos.restype = llama_token
# LLAMA_API llama_token llama_token_nl (const struct llama_context * ctx); // next-line
2023-09-29 02:42:03 +00:00
def llama_token_nl(ctx: llama_context_p) -> int:
2023-08-24 04:17:00 +00:00
return _lib.llama_token_nl(ctx)
_lib.llama_token_nl.argtypes = [llama_context_p]
_lib.llama_token_nl.restype = llama_token
# //
# // Tokenization
# //
2023-09-29 02:42:03 +00:00
# // Convert the provided text into tokens.
# // The tokens pointer must be large enough to hold the resulting tokens.
# // Returns the number of tokens on success, no more than n_max_tokens
# // Returns a negative number on failure - the number of tokens that would have been returned
2023-08-24 04:17:00 +00:00
# LLAMA_API int llama_tokenize(
# const struct llama_model * model,
# const char * text,
2023-09-18 20:11:27 +00:00
# int text_len,
2023-08-24 04:17:00 +00:00
# llama_token * tokens,
# int n_max_tokens,
# bool add_bos);
2023-09-29 02:42:03 +00:00
def llama_tokenize(
2023-07-15 19:11:01 +00:00
model: llama_model_p,
2023-08-24 04:17:00 +00:00
text: bytes,
2023-09-18 20:11:27 +00:00
text_len: Union[c_int, int],
2023-08-24 04:17:00 +00:00
tokens, # type: Array[llama_token]
n_max_tokens: Union[c_int, int],
add_bos: Union[c_bool, bool],
2023-07-15 19:11:01 +00:00
) -> int:
2023-09-29 02:42:03 +00:00
return _lib.llama_tokenize(model, text, text_len, tokens, n_max_tokens, add_bos)
2023-07-15 19:11:01 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_tokenize.argtypes = [
2023-07-20 22:51:53 +00:00
llama_model_p,
2023-08-24 04:17:00 +00:00
c_char_p,
2023-09-18 20:11:27 +00:00
c_int,
2023-08-24 04:17:00 +00:00
llama_token_p,
2023-07-20 22:51:53 +00:00
c_int,
2023-08-24 04:17:00 +00:00
c_bool,
2023-07-20 22:51:53 +00:00
]
2023-09-29 02:42:03 +00:00
_lib.llama_tokenize.restype = c_int
2023-03-24 18:59:29 +00:00
2023-08-27 16:59:20 +00:00
# // Token Id -> Piece.
# // Uses the vocabulary in the provided context.
# // Does not write null terminator to the buffer.
# // User code is responsible to remove the leading whitespace of the first non-BOS token when decoding multiple tokens.
# LLAMA_API int llama_token_to_piece(
2023-09-29 02:42:03 +00:00
# const struct llama_model * model,
# llama_token token,
# char * buf,
# int length);
2023-08-27 16:59:20 +00:00
def llama_token_to_piece(
2023-09-29 02:42:03 +00:00
model: llama_model_p,
token: llama_token,
buf: Union[c_char_p, bytes],
length: Union[c_int, int],
2023-08-24 04:17:00 +00:00
) -> int:
2023-09-29 02:42:03 +00:00
return _lib.llama_token_to_piece(model, token, buf, length)
2023-03-24 18:59:29 +00:00
2023-09-29 02:42:03 +00:00
_lib.llama_token_to_piece.argtypes = [llama_model_p, llama_token, c_char_p, c_int]
2023-08-27 16:59:20 +00:00
_lib.llama_token_to_piece.restype = c_int
2023-03-24 18:59:29 +00:00
2023-08-24 04:17:00 +00:00
# //
2023-07-24 17:04:34 +00:00
# // Grammar
# //
2023-08-24 04:17:00 +00:00
2023-07-24 17:04:34 +00:00
# LLAMA_API struct llama_grammar * llama_grammar_init(
# const llama_grammar_element ** rules,
# size_t n_rules,
# size_t start_rule_index);
def llama_grammar_init(
rules, # type: Array[llama_grammar_element_p] # type: ignore
n_rules: Union[c_size_t, int],
start_rule_index: Union[c_size_t, int],
2023-07-24 17:04:34 +00:00
) -> llama_grammar_p:
return _lib.llama_grammar_init(rules, n_rules, start_rule_index)
_lib.llama_grammar_init.argtypes = [
POINTER(llama_grammar_element_p),
c_size_t,
c_size_t,
]
_lib.llama_grammar_init.restype = llama_grammar_p
# LLAMA_API void llama_grammar_free(struct llama_grammar * grammar);
def llama_grammar_free(grammar: llama_grammar_p):
return _lib.llama_grammar_free(grammar)
_lib.llama_grammar_free.argtypes = [llama_grammar_p]
_lib.llama_grammar_free.restype = None
2023-08-25 18:35:53 +00:00
2023-09-09 16:12:32 +00:00
# LLAMA_API struct llama_grammar * llama_grammar_copy(const struct llama_grammar * grammar);
def llama_grammar_copy(grammar: llama_grammar_p) -> llama_grammar_p:
return _lib.llama_grammar_copy(grammar)
2023-08-25 18:35:53 +00:00
2023-09-09 16:12:32 +00:00
_lib.llama_grammar_copy.argtypes = [llama_grammar_p]
_lib.llama_grammar_copy.restype = llama_grammar_p
2023-07-24 17:04:34 +00:00
2023-08-24 04:17:00 +00:00
# //
# // Sampling functions
# //
2023-05-01 18:02:06 +00:00
2023-09-29 02:42:03 +00:00
# // Sets the current rng seed.
# LLAMA_API void llama_set_rng_seed(struct llama_context * ctx, uint32_t seed);
def llama_set_rng_seed(ctx: llama_context_p, seed: Union[c_uint32, int]):
return _lib.llama_set_rng_seed(ctx, seed)
_lib.llama_set_rng_seed.argtypes = [llama_context_p, c_uint32]
_lib.llama_set_rng_seed.restype = None
2023-05-01 18:02:06 +00:00
# @details Repetition penalty described in CTRL academic paper https://arxiv.org/abs/1909.05858, with negative logit fix.
2023-09-29 02:42:03 +00:00
# LLAMA_API void llama_sample_repetition_penalty(
# struct llama_context * ctx,
# llama_token_data_array * candidates,
# const llama_token * last_tokens,
# size_t last_tokens_size,
# float penalty);
2023-05-01 14:44:28 +00:00
def llama_sample_repetition_penalty(
ctx: llama_context_p,
2023-05-05 18:12:26 +00:00
candidates, # type: _Pointer[llama_token_data_array]
last_tokens_data, # type: Array[llama_token]
last_tokens_size: Union[c_int, int],
penalty: Union[c_float, float],
2023-05-01 18:02:06 +00:00
):
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_repetition_penalty(
ctx, candidates, last_tokens_data, last_tokens_size, penalty
)
_lib.llama_sample_repetition_penalty.argtypes = [
llama_context_p,
llama_token_data_array_p,
llama_token_p,
c_int,
c_float,
]
2023-05-01 18:02:06 +00:00
_lib.llama_sample_repetition_penalty.restype = None
2023-05-01 14:44:28 +00:00
2023-05-01 18:02:06 +00:00
# @details Frequency and presence penalties described in OpenAI API https://platform.openai.com/docs/api-reference/parameter-details.
2023-09-29 02:42:03 +00:00
# LLAMA_API void llama_sample_frequency_and_presence_penalties(
# struct llama_context * ctx,
# llama_token_data_array * candidates,
# const llama_token * last_tokens,
# size_t last_tokens_size,
# float alpha_frequency,
# float alpha_presence);
2023-05-01 14:44:28 +00:00
def llama_sample_frequency_and_presence_penalties(
2023-03-24 18:35:41 +00:00
ctx: llama_context_p,
2023-05-05 18:12:26 +00:00
candidates, # type: _Pointer[llama_token_data_array]
last_tokens_data, # type: Array[llama_token]
last_tokens_size: Union[c_int, int],
alpha_frequency: Union[c_float, float],
alpha_presence: Union[c_float, float],
2023-05-01 18:02:06 +00:00
):
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_frequency_and_presence_penalties(
ctx,
candidates,
last_tokens_data,
last_tokens_size,
alpha_frequency,
alpha_presence,
2023-03-24 18:35:41 +00:00
)
2023-03-23 09:33:06 +00:00
2023-05-01 14:44:28 +00:00
_lib.llama_sample_frequency_and_presence_penalties.argtypes = [
llama_context_p,
2023-05-01 14:44:28 +00:00
llama_token_data_array_p,
llama_token_p,
c_int,
2023-05-01 14:44:28 +00:00
c_float,
c_float,
]
2023-05-01 18:02:06 +00:00
_lib.llama_sample_frequency_and_presence_penalties.restype = None
2023-05-01 14:44:28 +00:00
2023-07-15 19:11:01 +00:00
# /// @details Apply classifier-free guidance to the logits as described in academic paper "Stay on topic with Classifier-Free Guidance" https://arxiv.org/abs/2306.17806
# /// @param candidates A vector of `llama_token_data` containing the candidate tokens, the logits must be directly extracted from the original generation context without being sorted.
# /// @params guidance_ctx A separate context from the same model. Other than a negative prompt at the beginning, it should have all generated and user input tokens copied from the main context.
# /// @params scale Guidance strength. 1.0f means no guidance. Higher values mean stronger guidance.
# LLAMA_API void llama_sample_classifier_free_guidance(
# struct llama_context * ctx,
# llama_token_data_array * candidates,
# struct llama_context * guidance_ctx,
2023-07-21 16:33:27 +00:00
# float scale);
2023-07-15 19:11:01 +00:00
def llama_sample_classifier_free_guidance(
ctx: llama_context_p,
candidates, # type: _Pointer[llama_token_data_array]
guidance_ctx: llama_context_p,
scale: Union[c_float, float],
2023-07-15 19:11:01 +00:00
):
return _lib.llama_sample_classifier_free_guidance(
2023-07-21 16:33:27 +00:00
ctx, candidates, guidance_ctx, scale
2023-07-15 19:11:01 +00:00
)
_lib.llama_sample_classifier_free_guidance.argtypes = [
llama_context_p,
llama_token_data_array_p,
llama_context_p,
c_float,
]
_lib.llama_sample_classifier_free_guidance.restype = None
2023-05-01 18:02:06 +00:00
# @details Sorts candidate tokens by their logits in descending order and calculate probabilities based on logits.
2023-09-29 02:42:03 +00:00
# LLAMA_API void llama_sample_softmax(
# struct llama_context * ctx,
# llama_token_data_array * candidates);
2023-05-05 18:12:26 +00:00
def llama_sample_softmax(
ctx: llama_context_p, candidates # type: _Pointer[llama_token_data]
):
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_softmax(ctx, candidates)
_lib.llama_sample_softmax.argtypes = [
llama_context_p,
llama_token_data_array_p,
]
2023-05-01 18:02:06 +00:00
_lib.llama_sample_softmax.restype = None
2023-05-01 14:44:28 +00:00
2023-05-01 18:02:06 +00:00
# @details Top-K sampling described in academic paper "The Curious Case of Neural Text Degeneration" https://arxiv.org/abs/1904.09751
2023-09-29 02:42:03 +00:00
# LLAMA_API void llama_sample_top_k(
# struct llama_context * ctx,
# llama_token_data_array * candidates,
# int k,
# size_t min_keep);
2023-05-01 18:47:55 +00:00
def llama_sample_top_k(
2023-05-05 17:54:22 +00:00
ctx: llama_context_p,
2023-05-05 18:12:26 +00:00
candidates, # type: _Pointer[llama_token_data_array]
k: Union[c_int, int],
min_keep: Union[c_size_t, int],
2023-05-01 18:47:55 +00:00
):
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_top_k(ctx, candidates, k, min_keep)
_lib.llama_sample_top_k.argtypes = [
llama_context_p,
llama_token_data_array_p,
c_int,
2023-05-01 18:47:55 +00:00
c_size_t,
2023-05-01 14:44:28 +00:00
]
2023-05-01 18:02:06 +00:00
_lib.llama_sample_top_k.restype = None
2023-05-01 14:44:28 +00:00
2023-05-01 18:02:06 +00:00
# @details Nucleus sampling described in academic paper "The Curious Case of Neural Text Degeneration" https://arxiv.org/abs/1904.09751
2023-09-29 02:42:03 +00:00
# LLAMA_API void llama_sample_top_p(
# struct llama_context * ctx,
# llama_token_data_array * candidates,
# float p,
# size_t min_keep);
2023-05-01 18:47:55 +00:00
def llama_sample_top_p(
2023-05-05 17:54:22 +00:00
ctx: llama_context_p,
2023-05-05 18:12:26 +00:00
candidates, # type: _Pointer[llama_token_data_array]
p: Union[c_float, float],
min_keep: Union[c_size_t, int],
2023-05-01 18:47:55 +00:00
):
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_top_p(ctx, candidates, p, min_keep)
_lib.llama_sample_top_p.argtypes = [
llama_context_p,
llama_token_data_array_p,
c_float,
2023-05-01 18:47:55 +00:00
c_size_t,
2023-05-01 14:44:28 +00:00
]
2023-05-01 18:02:06 +00:00
_lib.llama_sample_top_p.restype = None
2023-05-01 14:44:28 +00:00
2023-05-01 18:02:06 +00:00
# @details Tail Free Sampling described in https://www.trentonbricken.com/Tail-Free-Sampling/.
2023-09-29 02:42:03 +00:00
# LLAMA_API void llama_sample_tail_free(
# struct llama_context * ctx,
# llama_token_data_array * candidates,
# float z,
# size_t min_keep);
2023-05-01 14:44:28 +00:00
def llama_sample_tail_free(
ctx: llama_context_p,
2023-05-05 18:12:26 +00:00
candidates, # type: _Pointer[llama_token_data_array]
z: Union[c_float, float],
min_keep: Union[c_size_t, int],
2023-05-01 18:02:06 +00:00
):
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_tail_free(ctx, candidates, z, min_keep)
_lib.llama_sample_tail_free.argtypes = [
llama_context_p,
llama_token_data_array_p,
c_float,
2023-05-01 18:47:55 +00:00
c_size_t,
2023-05-01 14:44:28 +00:00
]
2023-05-01 18:02:06 +00:00
_lib.llama_sample_tail_free.restype = None
2023-05-01 14:44:28 +00:00
2023-05-01 18:02:06 +00:00
# @details Locally Typical Sampling implementation described in the paper https://arxiv.org/abs/2202.00666.
2023-09-29 02:42:03 +00:00
# LLAMA_API void llama_sample_typical(
# struct llama_context * ctx,
# llama_token_data_array * candidates,
# float p,
# size_t min_keep);
2023-05-01 18:47:55 +00:00
def llama_sample_typical(
ctx: llama_context_p,
2023-05-05 18:12:26 +00:00
candidates, # type: _Pointer[llama_token_data_array]
p: Union[c_float, float],
min_keep: Union[c_size_t, int],
2023-05-01 18:47:55 +00:00
):
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_typical(ctx, candidates, p, min_keep)
_lib.llama_sample_typical.argtypes = [
llama_context_p,
llama_token_data_array_p,
c_float,
2023-05-01 18:47:55 +00:00
c_size_t,
2023-05-01 14:44:28 +00:00
]
2023-05-01 18:02:06 +00:00
_lib.llama_sample_typical.restype = None
2023-05-01 14:44:28 +00:00
2023-09-29 02:42:03 +00:00
# LLAMA_API void llama_sample_temp(
# struct llama_context * ctx,
# llama_token_data_array * candidates,
# float temp);
def llama_sample_temp(
ctx: llama_context_p,
candidates, # type: _Pointer[llama_token_data_array]
temp: Union[c_float, float],
):
return _lib.llama_sample_temp(ctx, candidates, temp)
_lib.llama_sample_temp.argtypes = [
llama_context_p,
llama_token_data_array_p,
c_float,
]
_lib.llama_sample_temp.restype = None
# LLAMA_API DEPRECATED(void llama_sample_temperature(
# struct llama_context * ctx,
# llama_token_data_array * candidates,
# float temp),
# "use llama_sample_temp instead");
def llama_sample_temperature(
2023-05-05 18:12:26 +00:00
ctx: llama_context_p,
candidates, # type: _Pointer[llama_token_data_array]
temp: Union[c_float, float],
):
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_temperature(ctx, candidates, temp)
_lib.llama_sample_temperature.argtypes = [
llama_context_p,
llama_token_data_array_p,
2023-03-29 01:10:23 +00:00
c_float,
2023-05-01 14:44:28 +00:00
]
2023-05-01 18:02:06 +00:00
_lib.llama_sample_temperature.restype = None
2023-05-01 14:44:28 +00:00
2023-07-24 19:42:31 +00:00
# LLAMA_API void llama_sample_grammar(struct llama_context * ctx, llama_token_data_array * candidates, const struct llama_grammar * grammar);
def llama_sample_grammar(
ctx: llama_context_p,
candidates, # type: _Pointer[llama_token_data_array]
2023-08-05 05:43:35 +00:00
grammar, # type: llama_grammar_p
2023-07-24 19:42:31 +00:00
):
return _lib.llama_sample_grammar(ctx, candidates, grammar)
_lib.llama_sample_grammar.argtypes = [
llama_context_p,
llama_token_data_array_p,
llama_grammar_p,
]
_lib.llama_sample_grammar.restype = None
2023-05-01 18:02:06 +00:00
# @details Mirostat 1.0 algorithm described in the paper https://arxiv.org/abs/2007.14966. Uses tokens instead of words.
# @param candidates A vector of `llama_token_data` containing the candidate tokens, their probabilities (p), and log-odds (logit) for the current position in the generated text.
# @param tau The target cross-entropy (or surprise) value you want to achieve for the generated text. A higher value corresponds to more surprising or less predictable text, while a lower value corresponds to less surprising or more predictable text.
# @param eta The learning rate used to update `mu` based on the error between the target and observed surprisal of the sampled word. A larger learning rate will cause `mu` to be updated more quickly, while a smaller learning rate will result in slower updates.
# @param m The number of tokens considered in the estimation of `s_hat`. This is an arbitrary value that is used to calculate `s_hat`, which in turn helps to calculate the value of `k`. In the paper, they use `m = 100`, but you can experiment with different values to see how it affects the performance of the algorithm.
# @param mu Maximum cross-entropy. This value is initialized to be twice the target cross-entropy (`2 * tau`) and is updated in the algorithm based on the error between the target and observed surprisal.
2023-09-29 02:42:03 +00:00
# LLAMA_API llama_token llama_sample_token_mirostat(
# struct llama_context * ctx,
# llama_token_data_array * candidates,
# float tau,
# float eta,
# int m,
# float * mu);
2023-05-01 14:44:28 +00:00
def llama_sample_token_mirostat(
ctx: llama_context_p,
2023-05-05 18:12:26 +00:00
candidates, # type: _Pointer[llama_token_data_array]
tau: Union[c_float, float],
eta: Union[c_float, float],
m: Union[c_int, int],
2023-05-05 18:12:26 +00:00
mu, # type: _Pointer[c_float]
) -> int:
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_token_mirostat(ctx, candidates, tau, eta, m, mu)
_lib.llama_sample_token_mirostat.argtypes = [
llama_context_p,
llama_token_data_array_p,
c_float,
c_float,
c_int,
2023-05-05 18:04:12 +00:00
c_float_p,
2023-05-01 14:44:28 +00:00
]
_lib.llama_sample_token_mirostat.restype = llama_token
2023-05-01 18:02:06 +00:00
# @details Mirostat 2.0 algorithm described in the paper https://arxiv.org/abs/2007.14966. Uses tokens instead of words.
# @param candidates A vector of `llama_token_data` containing the candidate tokens, their probabilities (p), and log-odds (logit) for the current position in the generated text.
# @param tau The target cross-entropy (or surprise) value you want to achieve for the generated text. A higher value corresponds to more surprising or less predictable text, while a lower value corresponds to less surprising or more predictable text.
# @param eta The learning rate used to update `mu` based on the error between the target and observed surprisal of the sampled word. A larger learning rate will cause `mu` to be updated more quickly, while a smaller learning rate will result in slower updates.
# @param mu Maximum cross-entropy. This value is initialized to be twice the target cross-entropy (`2 * tau`) and is updated in the algorithm based on the error between the target and observed surprisal.
2023-09-29 02:42:03 +00:00
# LLAMA_API llama_token llama_sample_token_mirostat_v2(
# struct llama_context * ctx,
# llama_token_data_array * candidates,
# float tau,
# float eta,
# float * mu);
2023-05-01 14:44:28 +00:00
def llama_sample_token_mirostat_v2(
ctx: llama_context_p,
2023-05-05 18:12:26 +00:00
candidates, # type: _Pointer[llama_token_data_array]
tau: Union[c_float, float],
eta: Union[c_float, float],
2023-05-05 18:12:26 +00:00
mu, # type: _Pointer[c_float]
) -> int:
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_token_mirostat_v2(ctx, candidates, tau, eta, mu)
_lib.llama_sample_token_mirostat_v2.argtypes = [
llama_context_p,
llama_token_data_array_p,
2023-03-29 01:10:23 +00:00
c_float,
c_float,
2023-05-05 18:04:12 +00:00
c_float_p,
2023-05-01 14:44:28 +00:00
]
_lib.llama_sample_token_mirostat_v2.restype = llama_token
2023-05-01 18:02:06 +00:00
# @details Selects the token with the highest probability.
2023-09-29 02:42:03 +00:00
# LLAMA_API llama_token llama_sample_token_greedy(
# struct llama_context * ctx,
# llama_token_data_array * candidates);
def llama_sample_token_greedy(
2023-05-05 18:12:26 +00:00
ctx: llama_context_p,
candidates, # type: _Pointer[llama_token_data_array]
) -> int:
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_token_greedy(ctx, candidates)
_lib.llama_sample_token_greedy.argtypes = [
llama_context_p,
llama_token_data_array_p,
]
_lib.llama_sample_token_greedy.restype = llama_token
2023-05-01 18:02:06 +00:00
# @details Randomly selects a token from the candidates based on their probabilities.
2023-09-29 02:42:03 +00:00
# LLAMA_API llama_token llama_sample_token(
# struct llama_context * ctx,
# llama_token_data_array * candidates);
def llama_sample_token(
2023-05-05 18:12:26 +00:00
ctx: llama_context_p,
candidates, # type: _Pointer[llama_token_data_array]
) -> int:
2023-05-01 14:44:28 +00:00
return _lib.llama_sample_token(ctx, candidates)
_lib.llama_sample_token.argtypes = [
llama_context_p,
llama_token_data_array_p,
]
2023-05-01 14:44:28 +00:00
_lib.llama_sample_token.restype = llama_token
2023-07-24 19:55:26 +00:00
# /// @details Accepts the sampled token into the grammar
2023-09-29 02:42:03 +00:00
# LLAMA_API void llama_grammar_accept_token(
# struct llama_context * ctx,
# struct llama_grammar * grammar,
# llama_token token);
2023-07-24 19:55:26 +00:00
def llama_grammar_accept_token(
2023-08-05 05:43:35 +00:00
ctx: llama_context_p,
grammar: llama_grammar_p,
token: llama_token,
) -> None:
_lib.llama_grammar_accept_token(ctx, grammar, token)
2023-07-24 19:55:26 +00:00
_lib.llama_grammar_accept_token.argtypes = [
llama_context_p,
llama_grammar_p,
llama_token,
]
_lib.llama_grammar_accept_token.restype = None
2023-09-29 02:42:03 +00:00
2023-09-09 16:12:32 +00:00
# //
# // Beam search
# //
# struct llama_beam_view {
# const llama_token * tokens;
2023-09-29 02:42:03 +00:00
2023-09-09 16:12:32 +00:00
# size_t n_tokens;
2023-09-29 02:42:03 +00:00
# float p; // Cumulative beam probability (renormalized relative to all beams)
# bool eob; // Callback should set this to true when a beam is at end-of-beam.
2023-09-09 16:12:32 +00:00
# };
class llama_beam_view(ctypes.Structure):
_fields_ = [
("tokens", llama_token_p),
("n_tokens", c_size_t),
("p", c_float),
("eob", c_bool),
]
# // Passed to beam_search_callback function.
# // Whenever 0 < common_prefix_length, this number of tokens should be copied from any of the beams
# // (e.g. beams[0]) as they will be removed (shifted) from all beams in all subsequent callbacks.
# // These pointers are valid only during the synchronous callback, so should not be saved.
# struct llama_beams_state {
# struct llama_beam_view * beam_views;
# size_t n_beams; // Number of elements in beam_views[].
# size_t common_prefix_length; // Current max length of prefix tokens shared by all beams.
2023-09-29 02:42:03 +00:00
# bool last_call; // True iff this is the last callback invocation.
2023-09-09 16:12:32 +00:00
# };
class llama_beams_state(ctypes.Structure):
_fields_ = [
("beam_views", POINTER(llama_beam_view)),
("n_beams", c_size_t),
("common_prefix_length", c_size_t),
("last_call", c_bool),
]
# // Type of pointer to the beam_search_callback function.
# // void* callback_data is any custom data passed to llama_beam_search, that is subsequently
# // passed back to beam_search_callback. This avoids having to use global variables in the callback.
# typedef void (*llama_beam_search_callback_fn_t)(void * callback_data, struct llama_beams_state);
llama_beam_search_callback_fn_t = ctypes.CFUNCTYPE(None, c_void_p, llama_beams_state)
# /// @details Deterministically returns entire sentence constructed by a beam search.
# /// @param ctx Pointer to the llama_context.
# /// @param callback Invoked for each iteration of the beam_search loop, passing in beams_state.
# /// @param callback_data A pointer that is simply passed back to callback.
# /// @param n_beams Number of beams to use.
# /// @param n_past Number of tokens already evaluated.
# /// @param n_predict Maximum number of tokens to predict. EOS may occur earlier.
# /// @param n_threads Number of threads as passed to llama_eval().
2023-09-29 02:42:03 +00:00
# LLAMA_API void llama_beam_search(
# struct llama_context * ctx,
# llama_beam_search_callback_fn_t callback,
# void * callback_data,
# size_t n_beams,
# int n_past,
# int n_predict);
2023-09-09 16:12:32 +00:00
def llama_beam_search(
ctx: llama_context_p,
callback: "ctypes._CFuncPtr[None, c_void_p, llama_beams_state]", # type: ignore
callback_data: c_void_p,
n_beams: Union[c_size_t, int],
n_past: Union[c_int, int],
n_predict: Union[c_int, int],
2023-09-09 16:12:32 +00:00
):
return _lib.llama_beam_search(
2023-09-29 02:42:03 +00:00
ctx, callback, callback_data, n_beams, n_past, n_predict
2023-09-09 16:12:32 +00:00
)
2023-09-29 02:42:03 +00:00
_lib.llama_beam_search.argtypes = [
llama_context_p,
llama_beam_search_callback_fn_t,
c_void_p,
c_size_t,
c_int,
c_int,
]
_lib.llama_beam_search.restype = None
2023-07-24 19:55:26 +00:00
# Performance information
2023-03-24 18:59:29 +00:00
2023-07-06 21:57:56 +00:00
# LLAMA_API struct llama_timings llama_get_timings(struct llama_context * ctx);
def llama_get_timings(ctx: llama_context_p) -> llama_timings:
return _lib.llama_get_timings(ctx)
_lib.llama_get_timings.argtypes = [llama_context_p]
_lib.llama_get_timings.restype = llama_timings
2023-05-21 21:47:21 +00:00
# LLAMA_API void llama_print_timings(struct llama_context * ctx);
2023-03-23 09:33:06 +00:00
def llama_print_timings(ctx: llama_context_p):
_lib.llama_print_timings(ctx)
2023-03-23 09:33:06 +00:00
2023-03-24 18:35:41 +00:00
_lib.llama_print_timings.argtypes = [llama_context_p]
_lib.llama_print_timings.restype = None
2023-05-21 21:47:21 +00:00
# LLAMA_API void llama_reset_timings(struct llama_context * ctx);
2023-03-23 09:33:06 +00:00
def llama_reset_timings(ctx: llama_context_p):
_lib.llama_reset_timings(ctx)
2023-03-23 09:33:06 +00:00
2023-03-24 18:35:41 +00:00
_lib.llama_reset_timings.argtypes = [llama_context_p]
_lib.llama_reset_timings.restype = None
# Print system information
2023-05-21 21:47:21 +00:00
# LLAMA_API const char * llama_print_system_info(void);
2023-03-23 09:33:06 +00:00
def llama_print_system_info() -> bytes:
return _lib.llama_print_system_info()
_lib.llama_print_system_info.argtypes = []
_lib.llama_print_system_info.restype = c_char_p
2023-05-21 21:47:21 +00:00
2023-08-24 04:17:00 +00:00
2023-09-29 02:42:03 +00:00
# NOTE: THIS IS CURRENTLY BROKEN AS ggml_log_callback IS NOT EXPOSED IN LLAMA.H
2023-08-24 04:17:00 +00:00
# // Set callback for all future logging events.
# // If this is not called, or NULL is supplied, everything is output on stderr.
2023-09-29 02:42:03 +00:00
# LLAMA_API void llama_log_set(ggml_log_callback log_callback, void * user_data);
2023-08-24 04:17:00 +00:00
def llama_log_set(
log_callback: "ctypes._FuncPointer", user_data: c_void_p # type: ignore
):
return _lib.llama_log_set(log_callback, user_data)
_lib.llama_log_set.argtypes = [llama_log_callback, c_void_p]
_lib.llama_log_set.restype = None
2023-08-29 11:36:20 +00:00
# LLAMA_API void llama_dump_timing_info_yaml(FILE * stream, const struct llama_context * ctx);
def llama_dump_timing_info_yaml(stream: ctypes.c_void_p, ctx: llama_context_p):
return _lib.llama_dump_timing_info_yaml(stream, ctx)
2023-09-09 16:12:32 +00:00
2023-08-29 11:36:20 +00:00
_lib.llama_dump_timing_info_yaml.argtypes = [ctypes.c_void_p, llama_context_p]
_lib.llama_dump_timing_info_yaml.restype = None