vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

better label (+1 squashed commits)

Browse source 
Squashed commits:

[f573b2c] cuda 3 target arch
This commit is contained in:
Concedo 2023-07-24 22:46:53 +08:00
parent 7555dae4cc
commit d8d2449bfb
2 changed files with 2 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

2
CMakeLists.txt
View file

@ -99,7 +99,7 @@ if (LLAMA_CUBLAS)
if (LLAMA_CUDA_DMMV_F16) if (LLAMA_CUDA_DMMV_F16)
set(CMAKE_CUDA_ARCHITECTURES "60;61") # needed for f16 CUDA intrinsics set(CMAKE_CUDA_ARCHITECTURES "60;61") # needed for f16 CUDA intrinsics
else() else()
set(CMAKE_CUDA_ARCHITECTURES "37;61") # lowest CUDA 12 standard + lowest for integer intrinsics set(CMAKE_CUDA_ARCHITECTURES "37;52;61") # lowest CUDA 12 standard + lowest for integer intrinsics
endif() endif()
endif() endif()
message(STATUS "Using CUDA architectures: ${CMAKE_CUDA_ARCHITECTURES}") message(STATUS "Using CUDA architectures: ${CMAKE_CUDA_ARCHITECTURES}")

2
koboldcpp.py
View file

@ -1642,7 +1642,7 @@ if __name__ == '__main__':
parser.add_argument("--noavx2", help="Do not use AVX2 instructions, a slower compatibility mode for older devices. Does not work with --clblast.", action='store_true') parser.add_argument("--noavx2", help="Do not use AVX2 instructions, a slower compatibility mode for older devices. Does not work with --clblast.", action='store_true')
parser.add_argument("--debugmode", help="Shows additional debug info in the terminal.", action='store_const', const=1, default=0) parser.add_argument("--debugmode", help="Shows additional debug info in the terminal.", action='store_const', const=1, default=0)
parser.add_argument("--skiplauncher", help="Doesn't display or use the new GUI launcher.", action='store_true') parser.add_argument("--skiplauncher", help="Doesn't display or use the new GUI launcher.", action='store_true')
parser.add_argument("--hordeconfig", help="Sets the display model name to something else, for easy use on AI Horde. Optional additional parameters set the horde max genlength, max ctxlen, API key and worker name.",metavar=('[hordemodelname]', '[hordelength] [hordemaxctx] [hordeapikey] [hordeworkername]'), nargs='+') parser.add_argument("--hordeconfig", help="Sets the display model name to something else, for easy use on AI Horde. Optional additional parameters set the horde max genlength, max ctxlen, API key and worker name.",metavar=('[hordemodelname]', '[hordegenlength] [hordemaxctx] [hordeapikey] [hordeworkername]'), nargs='+')
compatgroup = parser.add_mutually_exclusive_group() compatgroup = parser.add_mutually_exclusive_group()
compatgroup.add_argument("--noblas", help="Do not use OpenBLAS for accelerated prompt ingestion", action='store_true') compatgroup.add_argument("--noblas", help="Do not use OpenBLAS for accelerated prompt ingestion", action='store_true')
compatgroup.add_argument("--useclblast", help="Use CLBlast for GPU Acceleration. Must specify exactly 2 arguments, platform ID and device ID (e.g. --useclblast 1 0).", type=int, choices=range(0,9), nargs=2) compatgroup.add_argument("--useclblast", help="Use CLBlast for GPU Acceleration. Must specify exactly 2 arguments, platform ID and device ID (e.g. --useclblast 1 0).", type=int, choices=range(0,9), nargs=2)