Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-06 13:01:21 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge commit '20455a4ad336e958cfe8f82efce2c46cd44c4fa3' into concedo_experimental

Browse Source 
# Conflicts:
#	common/CMakeLists.txt
#	ggml/src/ggml-hexagon/ggml-hexagon.cpp
#	ggml/src/ggml-hexagon/htp/CMakeLists.txt
#	ggml/src/ggml-hexagon/htp/htp-ctx.h
#	ggml/src/ggml-hexagon/htp/htp-ops.h
#	ggml/src/ggml-hexagon/htp/main.c
#	scripts/sync_vendor.py
#	tests/test-chat.cpp
#	tests/test-reasoning-budget.cpp
#	tests/test-save-load-state.cpp
#	tools/server/CMakeLists.txt
#	tools/server/README.md
This commit is contained in:
Concedo
2026-07-27 22:12:11 +08:00
parent b0596e6202 20455a4ad3
commit 90510f2b27
99 changed files with 5408 additions and 644 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-quant.cpp
+1
View File
@@ -1357,6 +1357,7 @@ llama_model * llama_quant_model_from_metadata(const llama_quant_model_desc * des
model->hparams.n_embd_head_k_full = desc->n_embd_head_k;
model->hparams.n_embd_head_v_full = desc->n_embd_head_v;
model->hparams.n_layer_all = desc->n_layer;
GGML_ASSERT(desc->n_layer > 0 && desc->n_layer <= LLAMA_MAX_LAYERS);
model->hparams.n_expert = desc->n_expert;
for (uint32_t i = 0; i < desc->n_layer; i++) {