upstream-archive/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-08-04 00:50:47 -05:00
Code Issues Packages Projects Releases Wiki Activity

llama : load MTP tensors only if they are really used (#26296)

Browse Source 
* llama : load MTP tensors only if they are really used

* llama : skip loading MTP (if not used) in remaining models that support MTP

---------

Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
This commit is contained in:
fairydreaming
2026-07-31 14:57:02 +02:00
committed by GitHub
parent 6f3c0a790b
commit 82dbc4f017
13 changed files with 65 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/llama-quant.cpp
View File

@@ -893,7 +893,7 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::
const llama_model_kv_override * kv_overrides = params->kv_overrides;
std::vector<std::string> splits = {};
llama_model_loader ml(/*metadata*/ nullptr, /*set_tensor_data*/ nullptr, /*set_tensor_data_ud*/ nullptr,
fname_inp, splits, /*file*/ nullptr, /*load_mode*/ load_mode, /*check_tensors*/ true, /*no_alloc*/ false, kv_overrides, nullptr);
fname_inp, splits, /*file*/ nullptr, /*load_mode*/ load_mode, /*check_tensors*/ true, /*no_alloc*/ false, /*load_mtp*/ true, kv_overrides, nullptr);
ml.init_mappings(false); // no prefetching
auto mparams = llama_model_default_params();