vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-04-28 03:30:20 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 3

fix for override tensors not passing correctly

Browse source
This commit is contained in:
Concedo 2025-11-28 13:03:40 +08:00
parent 7527f1eff0
commit 9a46faa1c3
1 changed files with 2 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

2
gpttype_adapter.cpp
View file

@ -2258,6 +2258,8 @@ ModelLoadResult gpttype_load_model(const load_model_inputs inputs, FileFormat in
std::vector<llama_model_kv_override> kvos; //ensure it keeps in scope until model is created
std::vector<llama_model_tensor_buft_override> tenos; //ensure it keeps in scope until model is created
std::vector<std::string> temp_tensor_names; //store temp tensor names to have mem references.
temp_tensor_names.reserve(32); //very important, prevents vector from reallocating
tenos.reserve(32);
if(inputs.moe_experts>0)
{
printf("\nOverriding number of experts to %d\n",inputs.moe_experts);