vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-10 09:04:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'master' into concedo_experimental

Browse source 
# Conflicts:
#	CMakeLists.txt
#	Makefile
#	build.zig
#	flake.lock
#	flake.nix
#	ggml.c
This commit is contained in:
Concedo 2023-10-31 20:44:04 +08:00
commit cc5b282350
24 changed files with 2609 additions and 2512 deletions
Download patch file Download diff file Expand all files Collapse all files

2
gpttype_adapter.cpp
View file

@ -963,7 +963,7 @@ ModelLoadResult gpttype_load_model(const load_model_inputs inputs, FileFormat in
//determine mem per token
std::vector<int> tmp = {1, 2, 3, 4};
llama_kv_cache_tokens_rm(llama_ctx_v4, -1, -1);
llama_kv_cache_clear(llama_ctx_v4);
auto er = llama_decode(llama_ctx_v4, llama_batch_get_one(tmp.data(), tmp.size(), 0, 0));
if(er!=0)
{