public_git/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/whisper.cpp.git synced 2026-10-18 21:55:18 +02:00
Code Issues Packages Projects Releases Wiki Activity

talk-llama : sync llama.cpp

Browse Source
This commit is contained in:
Georgi Gerganov
2026-02-27 20:57:58 +02:00
parent 4734056067
commit 84f8db71d8
41 changed files with 1734 additions and 2227 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/talk-llama/models/models.h
+66 -99
View File
@@ -1,23 +1,51 @@
#pragma once
#include "../llama-model.h"
#include "../llama-graph.h"
#include "llama-model.h"
#include "llama-graph.h"
// TODO: remove in follow-up PR - move to .cpp files
#include "../llama-memory-recurrent.h"
// note: almost all graphs require atleast sqrtf, so include cmath globally
#include <cmath>
struct llm_graph_context_mamba : public llm_graph_context {
llm_graph_context_mamba(const llm_graph_params & params);
//
// base classes
//
virtual ~llm_graph_context_mamba() = default;
struct llm_build_mamba_base : public llm_graph_context {
llm_build_mamba_base(const llm_graph_params & params);
virtual ~llm_build_mamba_base() = default;
ggml_tensor * build_mamba_layer(llm_graph_input_rs * inp, ggml_tensor * cur, const llama_model & model, const llama_ubatch & ubatch, int il);
ggml_tensor * build_mamba2_layer(llm_graph_input_rs * inp, ggml_tensor * cur, const llama_model & model, const llama_ubatch & ubatch, int il) const;
};
// Base class for RWKV-related models
struct llm_build_delta_net_base : public llm_graph_context {
llm_build_delta_net_base(const llm_graph_params & params);
virtual ~llm_build_delta_net_base() = default;
// returns pair of output and new state
std::pair<ggml_tensor *, ggml_tensor *> build_delta_net_chunking(
ggml_tensor * q,
ggml_tensor * k,
ggml_tensor * v,
ggml_tensor * g,
ggml_tensor * b,
ggml_tensor * s,
int il);
// returns pair of output and new state
std::pair<ggml_tensor *, ggml_tensor *> build_delta_net_autoregressive(
ggml_tensor * q,
ggml_tensor * k,
ggml_tensor * v,
ggml_tensor * g,
ggml_tensor * b,
ggml_tensor * s,
int il);
};
struct llm_build_rwkv6_base : public llm_graph_context {
const llama_model & model;
@@ -58,6 +86,10 @@ struct llm_build_rwkv7_base : public llm_graph_context {
int il) const;
};
//
// models
//
struct llm_build_afmoe : public llm_graph_context {
llm_build_afmoe(const llama_model & model, const llm_graph_params & params);
};
@@ -158,6 +190,10 @@ struct llm_build_ernie4_5_moe : public llm_graph_context {
llm_build_ernie4_5_moe(const llama_model & model, const llm_graph_params & params);
};
struct llm_build_paddleocr : public llm_graph_context {
llm_build_paddleocr(const llama_model & model, const llm_graph_params & params);
};
template <bool iswa>
struct llm_build_exaone4 : public llm_graph_context {
llm_build_exaone4(const llama_model & model, const llm_graph_params & params);
@@ -175,7 +211,7 @@ struct llm_build_falcon : public llm_graph_context {
llm_build_falcon(const llama_model & model, const llm_graph_params & params);
};
struct llm_build_falcon_h1 : public llm_graph_context_mamba {
struct llm_build_falcon_h1 : public llm_build_mamba_base {
llm_build_falcon_h1(const llama_model & model, const llm_graph_params & params);
};
@@ -253,7 +289,7 @@ private:
const int il);
};
struct llm_build_granite_hybrid : public llm_graph_context_mamba {
struct llm_build_granite_hybrid : public llm_build_mamba_base {
llm_build_granite_hybrid(const llama_model & model, const llm_graph_params & params);
ggml_tensor * build_layer_ffn(ggml_tensor * cur, ggml_tensor * inpSA, const llama_model & model, const int il);
ggml_tensor * build_attention_layer(ggml_tensor * cur, ggml_tensor * inp_pos, llm_graph_input_attn_kv * inp_attn,
@@ -284,11 +320,15 @@ struct llm_build_jais : public llm_graph_context {
llm_build_jais(const llama_model & model, const llm_graph_params & params);
};
struct llm_build_jamba : public llm_graph_context_mamba {
struct llm_build_jais2 : public llm_graph_context {
llm_build_jais2(const llama_model & model, const llm_graph_params & params);
};
struct llm_build_jamba : public llm_build_mamba_base {
llm_build_jamba(const llama_model & model, const llm_graph_params & params);
};
struct llm_build_kimi_linear : public llm_graph_context_mamba {
struct llm_build_kimi_linear : public llm_build_delta_net_base {
llm_build_kimi_linear(const llama_model & model, const llm_graph_params & params);
std::pair<ggml_tensor *, ggml_tensor *> build_kda_autoregressive(
@@ -315,15 +355,9 @@ struct llm_build_kimi_linear : public llm_graph_context_mamba {
const llama_model & model;
};
template <bool iswa>
struct llm_build_lfm2 : public llm_graph_context {
const llama_model & model;
llm_build_lfm2(const llama_model & model, const llm_graph_params & params);
ggml_tensor * build_moe_feed_forward(ggml_tensor * cur, int il) const;
ggml_tensor * build_dense_feed_forward(ggml_tensor * cur, int il) const;
ggml_tensor * build_attn_block(ggml_tensor * cur, ggml_tensor * inp_pos, llm_graph_input_attn_kv * inp_attn, int il) const;
ggml_tensor * build_shortconv_block(ggml_tensor * cur, llm_graph_input_rs * inp_recr, int il);
};
struct llm_build_llada : public llm_graph_context {
@@ -347,7 +381,7 @@ struct llm_build_maincoder : public llm_graph_context {
llm_build_maincoder(const llama_model & model, const llm_graph_params & params);
};
struct llm_build_mamba : public llm_graph_context_mamba {
struct llm_build_mamba : public llm_build_mamba_base {
llm_build_mamba(const llama_model & model, const llm_graph_params & params);
};
@@ -379,17 +413,21 @@ struct llm_build_nemotron : public llm_graph_context {
llm_build_nemotron(const llama_model & model, const llm_graph_params & params);
};
struct llm_build_nemotron_h : public llm_graph_context_mamba {
struct llm_build_nemotron_h : public llm_build_mamba_base {
llm_build_nemotron_h(const llama_model & model, const llm_graph_params & params);
ggml_tensor * build_ffn_layer(ggml_tensor * cur, const llama_model & model, const int il);
ggml_tensor * build_ffn_layer(ggml_tensor * cur, const llama_model & model, int il);
ggml_tensor * build_attention_layer(ggml_tensor * cur, llm_graph_input_attn_kv * inp_attn,
const llama_model & model, const int64_t n_embd_head, const int il);
const llama_model & model, int64_t n_embd_head, int il);
};
struct llm_build_neo_bert : public llm_graph_context {
llm_build_neo_bert(const llama_model & model, const llm_graph_params & params);
};
struct llm_build_eurobert : public llm_graph_context {
llm_build_eurobert(const llama_model & model, const llm_graph_params & params);
};
template <bool iswa>
struct llm_build_olmo2 : public llm_graph_context {
llm_build_olmo2(const llama_model & model, const llm_graph_params & params);
@@ -428,7 +466,7 @@ struct llm_build_phi3 : public llm_graph_context {
llm_build_phi3(const llama_model & model, const llm_graph_params & params);
};
struct llm_build_plamo2 : public llm_graph_context_mamba {
struct llm_build_plamo2 : public llm_build_mamba_base {
llm_build_plamo2(const llama_model & model, const llm_graph_params & params);
private:
ggml_tensor * build_plamo2_mamba_layer(llm_graph_input_rs * inp, ggml_tensor * cur, const llama_model & model, const llama_ubatch & ubatch, int il);
@@ -477,7 +515,7 @@ struct llm_build_qwen3vlmoe : public llm_graph_context {
llm_build_qwen3vlmoe(const llama_model & model, const llm_graph_params & params);
};
struct llm_build_qwen3next : public llm_graph_context_mamba {
struct llm_build_qwen3next : public llm_build_delta_net_base {
llm_build_qwen3next(const llama_model & model, const llm_graph_params & params);
private:
ggml_tensor * build_layer_attn(
@@ -495,26 +533,6 @@ private:
ggml_tensor * cur,
int il);
// returns pair of output and new state
std::pair<ggml_tensor *, ggml_tensor *> build_delta_net_chunking(
ggml_tensor * q,
ggml_tensor * k,
ggml_tensor * v,
ggml_tensor * g,
ggml_tensor * beta,
ggml_tensor * state,
int il);
// returns pair of output and new state
std::pair<ggml_tensor *, ggml_tensor *> build_delta_net_autoregressive(
ggml_tensor * q,
ggml_tensor * k,
ggml_tensor * v,
ggml_tensor * g,
ggml_tensor * beta,
ggml_tensor * state,
int il);
ggml_tensor * build_norm_gated(
ggml_tensor * input,
ggml_tensor * weights,
@@ -529,7 +547,7 @@ private:
const llama_model & model;
};
struct llm_build_qwen35 : public llm_graph_context_mamba {
struct llm_build_qwen35 : public llm_build_delta_net_base {
llm_build_qwen35(const llama_model & model, const llm_graph_params & params);
private:
ggml_tensor * build_layer_attn(
@@ -542,38 +560,12 @@ private:
ggml_tensor * build_layer_attn_linear(
llm_graph_input_rs * inp,
ggml_tensor * cur,
ggml_tensor * causal_mask,
ggml_tensor * identity,
ggml_tensor * diag_mask,
int il);
ggml_tensor * build_layer_ffn(
ggml_tensor * cur,
int il);
// returns pair of output and new state
std::pair<ggml_tensor *, ggml_tensor *> build_delta_net_chunking(
ggml_tensor * q,
ggml_tensor * k,
ggml_tensor * v,
ggml_tensor * g,
ggml_tensor * beta,
ggml_tensor * state,
ggml_tensor * causal_mask,
ggml_tensor * identity,
ggml_tensor * diag_mask,
int il);
// returns pair of output and new state
std::pair<ggml_tensor *, ggml_tensor *> build_delta_net_autoregressive(
ggml_tensor * q,
ggml_tensor * k,
ggml_tensor * v,
ggml_tensor * g,
ggml_tensor * beta,
ggml_tensor * state,
int il);
ggml_tensor * build_norm_gated(
ggml_tensor * input,
ggml_tensor * weights,
@@ -588,7 +580,8 @@ private:
const llama_model & model;
};
struct llm_build_qwen35moe : public llm_graph_context_mamba {
// TODO: derive llm_build_delta_net_base instead
struct llm_build_qwen35moe : public llm_build_delta_net_base {
llm_build_qwen35moe(const llama_model & model, const llm_graph_params & params);
private:
ggml_tensor * build_layer_attn(
@@ -601,38 +594,12 @@ private:
ggml_tensor * build_layer_attn_linear(
llm_graph_input_rs * inp,
ggml_tensor * cur,
ggml_tensor * causal_mask,
ggml_tensor * identity,
ggml_tensor * diag_mask,
int il);
ggml_tensor * build_layer_ffn(
ggml_tensor * cur,
int il);
// returns pair of output and new state
std::pair<ggml_tensor *, ggml_tensor *> build_delta_net_chunking(
ggml_tensor * q,
ggml_tensor * k,
ggml_tensor * v,
ggml_tensor * g,
ggml_tensor * beta,
ggml_tensor * state,
ggml_tensor * causal_mask,
ggml_tensor * identity,
ggml_tensor * diag_mask,
int il);
// returns pair of output and new state
std::pair<ggml_tensor *, ggml_tensor *> build_delta_net_autoregressive(
ggml_tensor * q,
ggml_tensor * k,
ggml_tensor * v,
ggml_tensor * g,
ggml_tensor * beta,
ggml_tensor * state,
int il);
ggml_tensor * build_norm_gated(
ggml_tensor * input,
ggml_tensor * weights,