thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

llama : re-enable manual LoRA adapter free (#19983)

Browse source 
* Re-enable manual LoRA adapter free

* Remove stale "all adapters must be loaded before context creation" stale comments
This commit is contained in:
Pop Flamingo 2026-03-18 11:03:26 +01:00 • committed by GitHub
parent f4049ad735
commit 312cf03328
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
5 changed files with 19 additions and 12 deletions
Download patch file Download diff file Expand all files Collapse all files

15
src/llama-adapter.cpp
View file

@ -418,7 +418,7 @@ static void llama_adapter_lora_init_impl(llama_model & model, const char * path_
}
llama_adapter_lora * llama_adapter_lora_init(llama_model * model, const char * path_lora) {
llama_adapter_lora * adapter = new llama_adapter_lora();
llama_adapter_lora * adapter = new llama_adapter_lora(model);
try {
llama_adapter_lora_init_impl(*model, path_lora, *adapter);
@ -471,8 +471,17 @@ int32_t llama_adapter_meta_val_str_by_index(const llama_adapter_lora * adapter,
return snprintf(buf, buf_size, "%s", it->second.c_str());
}
void llama_adapter_lora_free(llama_adapter_lora *) {
// deprecated: adapters are freed by llama_model's destructor
void llama_adapter_lora_free(llama_adapter_lora * adapter) {
if (adapter == nullptr) {
return;
}
if (adapter->model != nullptr) {
adapter->model->loras.erase(adapter);
adapter->model = nullptr;
}
delete adapter;
}
uint64_t llama_adapter_get_alora_n_invocation_tokens(const struct llama_adapter_lora * adapter) {

4
src/llama-adapter.h
View file

@ -61,6 +61,8 @@ struct llama_adapter_lora_weight {
};
struct llama_adapter_lora {
llama_model * model = nullptr;
// map tensor name to lora_a_b
std::unordered_map<std::string, llama_adapter_lora_weight> ab_map;
@ -75,7 +77,7 @@ struct llama_adapter_lora {
// activated lora (aLoRA)
std::vector<llama_token> alora_invocation_tokens;
llama_adapter_lora() = default;
explicit llama_adapter_lora(llama_model * model) : model(model) {}
~llama_adapter_lora() = default;
llama_adapter_lora_weight * get_weight(ggml_tensor * w);