Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

llama: model_loader: add TENSOR_READ_LAZY (#27794)

Browse Source 
* llama: model_loader: add TENSOR_GET_ROW_LAZY

* add --tensor-read-lazy

* rename to TENSOR_READ_LAZY

* gen docs

* address comments
This commit is contained in:
Xuan-Son Nguyen
2026-08-27 15:14:34 +02:00
committed by GitHub
parent cae63579b6
commit fac889fb38
15 changed files with 129 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-model.cpp
+3 -1
View File
@@ -2631,6 +2631,7 @@ llama_model_params llama_model_default_params() {
/*.n_gpu_layers =*/ -1,
/*.split_mode =*/ LLAMA_SPLIT_MODE_LAYER,
/*.load_mode =*/ LLAMA_LOAD_MODE_AUTO,
/*.tensor_read_lazy =*/ LLAMA_TENSOR_READ_LAZY_AUTO,
/*.main_gpu =*/ 0,
/*.tensor_split =*/ nullptr,
/*.progress_callback =*/ nullptr,
@@ -3067,7 +3068,8 @@ llama_model_base::llama_model_base(const struct llama_model_params & params) : l
TENSOR_NOT_REQUIRED (llama_model_loader::TENSOR_NOT_REQUIRED),
TENSOR_SKIP (llama_model_loader::TENSOR_SKIP),
TENSOR_SKIP_IF_VIRTUAL(llama_model_loader::TENSOR_SKIP_IF_VIRTUAL),
TENSOR_ALLOW_RESHAPE (llama_model_loader::TENSOR_ALLOW_RESHAPE) {}
TENSOR_ALLOW_RESHAPE (llama_model_loader::TENSOR_ALLOW_RESHAPE),
TENSOR_READ_LAZY (llama_model_loader::TENSOR_READ_LAZY) {}
ggml_tensor * llama_model_base::create_tensor(const LLM_TN_IMPL & tn, const std::initializer_list<int64_t> & ne, int flags) {
GGML_ASSERT(ml != nullptr);