vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-11 09:34:37 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'master' into concedo_experimental

Browse source 
# Conflicts:
#	.github/workflows/build.yml
#	.github/workflows/docker.yml
#	CMakeLists.txt
#	Makefile
#	README.md
#	ci/README.md
#	ci/run.sh
#	flake.lock
#	ggml-metal.m
#	ggml-opencl.cpp
#	ggml-vulkan-shaders.hpp
#	ggml-vulkan.cpp
#	ggml-vulkan.h
#	ggml.c
#	ggml_vk_generate_shaders.py
#	llama.cpp
#	llama.h
#	pocs/vdot/vdot.cpp
#	tests/test-llama-grammar.cpp
#	tests/test-sampling.cpp
This commit is contained in:
Concedo 2024-01-29 23:12:09 +08:00
commit f73de33f74
37 changed files with 37432 additions and 21072 deletions
Download patch file Download diff file Expand all files Collapse all files

5
llama.h
View file

@ -6,6 +6,9 @@
#ifdef GGML_USE_CUBLAS
#include "ggml-cuda.h"
#define LLAMA_MAX_DEVICES 16
#elif defined(GGML_USE_SYCL)
#include "ggml-sycl.h"
#define LLAMA_MAX_DEVICES GGML_SYCL_MAX_DEVICES
#else
//just max it out, same as GGML_CUDA_MAX_DEVICES
#define LLAMA_MAX_DEVICES 16
@ -47,7 +50,7 @@
#define LLAMA_SESSION_MAGIC LLAMA_FILE_MAGIC_GGSN
#define LLAMA_SESSION_VERSION 4
#if defined(GGML_USE_CUBLAS) || defined(GGML_USE_CLBLAST) || defined(GGML_USE_METAL) || defined(GGML_USE_VULKAN)
#if defined(GGML_USE_CUBLAS) || defined(GGML_USE_CLBLAST) || defined(GGML_USE_METAL) || defined(GGML_USE_VULKAN) || defined(GGML_USE_SYCL)
// Defined when llama.cpp is compiled with support for offloading model layers to GPU.
#define LLAMA_SUPPORTS_GPU_OFFLOAD
#endif