vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-13 18:39:48 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

Merge branch 'master' into concedo_experimental

Browse source 
should fix multigpu
This commit is contained in:
Concedo 2023-11-05 22:14:40 +08:00
commit a62468ec4c
4 changed files with 158 additions and 86 deletions
Download patch file Download diff file Expand all files Collapse all files

11
llama.cpp
View file

@ -5196,11 +5196,12 @@ static int llama_decode_internal(
// If all tensors can be run on the GPU then using more than 1 thread is detrimental.
const bool full_offload_supported =
model.arch == LLM_ARCH_LLAMA ||
model.arch == LLM_ARCH_BAICHUAN ||
model.arch == LLM_ARCH_FALCON ||
model.arch == LLM_ARCH_REFACT ||
model.arch == LLM_ARCH_MPT;
model.arch == LLM_ARCH_LLAMA ||
model.arch == LLM_ARCH_BAICHUAN ||
model.arch == LLM_ARCH_FALCON ||
model.arch == LLM_ARCH_REFACT ||
model.arch == LLM_ARCH_MPT ||
model.arch == LLM_ARCH_STARCODER;
const bool fully_offloaded = model.n_gpu_layers >= (int) hparams.n_layer + 3;
if (ggml_cpu_has_cublas() && full_offload_supported && fully_offloaded) {