extern/whisper.cpp
1
0
Fork 1
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-08-13 23:58:35 +02:00
Code Issues Packages Projects Releases Wiki Activity

talk-llama : sync llama.cpp

Browse Source
This commit is contained in:
Georgi Gerganov
2025-07-28 10:09:47 +03:00
parent 5b4646df1a
commit d0a9d8c7f8
27 changed files with 3566 additions and 995 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
examples/talk-llama/llama.h
View File

@ -71,12 +71,13 @@ extern "C" {
typedef int32_t llama_seq_id;
enum llama_vocab_type {
LLAMA_VOCAB_TYPE_NONE = 0, // For models without vocab
LLAMA_VOCAB_TYPE_SPM = 1, // LLaMA tokenizer based on byte-level BPE with byte fallback
LLAMA_VOCAB_TYPE_BPE = 2, // GPT-2 tokenizer based on byte-level BPE
LLAMA_VOCAB_TYPE_WPM = 3, // BERT tokenizer based on WordPiece
LLAMA_VOCAB_TYPE_UGM = 4, // T5 tokenizer based on Unigram
LLAMA_VOCAB_TYPE_RWKV = 5, // RWKV tokenizer based on greedy tokenization
LLAMA_VOCAB_TYPE_NONE = 0, // For models without vocab
LLAMA_VOCAB_TYPE_SPM = 1, // LLaMA tokenizer based on byte-level BPE with byte fallback
LLAMA_VOCAB_TYPE_BPE = 2, // GPT-2 tokenizer based on byte-level BPE
LLAMA_VOCAB_TYPE_WPM = 3, // BERT tokenizer based on WordPiece
LLAMA_VOCAB_TYPE_UGM = 4, // T5 tokenizer based on Unigram
LLAMA_VOCAB_TYPE_RWKV = 5, // RWKV tokenizer based on greedy tokenization
LLAMA_VOCAB_TYPE_PLAMO2 = 6, // PLaMo-2 tokenizer based on Aho-Corasick with dynamic programming
};
enum llama_rope_type {
@ -334,6 +335,9 @@ extern "C" {
bool swa_full; // use full-size SWA cache (https://github.com/ggml-org/llama.cpp/pull/13194#issuecomment-2868343055)
// NOTE: setting to false when n_seq_max > 1 can cause bad performance in some cases
// ref: https://github.com/ggml-org/llama.cpp/pull/13845#issuecomment-2924800573
bool kv_unified; // use a unified buffer across the input sequences when computing the attention
// try to disable when n_seq_max > 1 for improved performance when the sequences do not share a large prefix
// ref: https://github.com/ggml-org/llama.cpp/pull/14363
};
// model quantization parameters
@ -724,7 +728,7 @@ extern "C" {
// - lazily on next llama_decode()
// p0 < 0 : [0, p1]
// p1 < 0 : [p0, inf)
DEPRECATED(void llama_kv_self_seq_div(
DEPRECATED(LLAMA_API void llama_kv_self_seq_div(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_pos p0,
@ -952,6 +956,7 @@ extern "C" {
// in the order they have appeared in the batch.
// Rows: number of tokens for which llama_batch.logits[i] != 0
// Cols: n_vocab
// TODO: deprecate in favor of llama_get_logits_ith() (ref: https://github.com/ggml-org/llama.cpp/pull/14853#issuecomment-3113143522)
LLAMA_API float * llama_get_logits(struct llama_context * ctx);
// Logits for the ith token. For positive indices, Equivalent to:
@ -966,6 +971,7 @@ extern "C" {
// in the order they have appeared in the batch.
// shape: [n_outputs*n_embd]
// Otherwise, returns NULL.
// TODO: deprecate in favor of llama_get_embeddings_ith() (ref: https://github.com/ggml-org/llama.cpp/pull/14853#issuecomment-3113143522)
LLAMA_API float * llama_get_embeddings(struct llama_context * ctx);
// Get the embeddings for the ith token. For positive indices, Equivalent to:
@ -1004,6 +1010,7 @@ extern "C" {
LLAMA_API llama_token llama_vocab_sep(const struct llama_vocab * vocab); // sentence separator
LLAMA_API llama_token llama_vocab_nl (const struct llama_vocab * vocab); // next-line
LLAMA_API llama_token llama_vocab_pad(const struct llama_vocab * vocab); // padding
LLAMA_API llama_token llama_vocab_mask(const struct llama_vocab * vocab); // mask
LLAMA_API bool llama_vocab_get_add_bos(const struct llama_vocab * vocab);
LLAMA_API bool llama_vocab_get_add_eos(const struct llama_vocab * vocab);
@ -1389,6 +1396,7 @@ extern "C" {
int32_t n_p_eval;
int32_t n_eval;
int32_t n_reused; // number of times a ggml compute graph had been reused
};
struct llama_perf_sampler_data {