extern/whisper.cpp
1
0
Fork 1
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-08-09 10:35:51 +02:00
Code Issues Packages Projects Releases Wiki Activity

talk-llama : sync llama.cpp

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2025-05-13 13:20:19 +03:00
parent a14c89aefa
commit f890560575
25 changed files with 2847 additions and 1125 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
examples/talk-llama/llama-model.h
View File

@ -76,6 +76,7 @@ enum llm_type {
LLM_TYPE_236B,
LLM_TYPE_290B,
LLM_TYPE_314B,
LLM_TYPE_405B,
LLM_TYPE_671B,
LLM_TYPE_SMALL,
LLM_TYPE_MEDIUM,
@ -95,6 +96,8 @@ enum llm_type {
LLM_TYPE_235B_A22B,
};
std::string llama_rope_scaling_type_name(llama_rope_scaling_type rope_scaling_type);
struct llama_layer_posnet {
// resnet
struct ggml_tensor * norm1 = nullptr;
@ -395,8 +398,11 @@ struct llama_model {
const struct ggml_tensor * get_tensor(const char * name) const;
ggml_tensor * get_rope_factors(uint32_t n_ctx_per_seq, int il) const;
// note: can mutate `cparams`
// TODO: move this to new llm_arch_model_i interface
llama_memory_i * create_memory() const; // TODO: params
llama_memory_i * create_memory(const llama_memory_params & params, llama_cparams & cparams) const;
// TODO: move this to new llm_arch_model_i interface
llm_graph_result_ptr build_graph(