ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-04-30 19:31:48 +00:00
Code Issues Packages Projects Releases Wiki Activity

Do not allocate KV cache for unused layers

Browse Source
This commit is contained in:
Iwan Kawrakow
2025-10-20 08:35:25 +03:00
parent 1789de5994
commit 599c812f12
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/llama.cpp
View File

@@ -532,7 +532,7 @@ static bool llama_kv_cache_init(
const struct llama_hparams & hparams = model.hparams; const struct llama_hparams & hparams = model.hparams;
const int64_t n_layer = hparams.n_layer; const int64_t n_layer = hparams.n_layer - hparams.nextn_predict_layers;
cache.has_shift = false; cache.has_shift = false;