vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-10 17:14:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fixed cfg scale, but turns out it sucks. embedded aria2c into pyinstaller

Browse source
This commit is contained in:
Concedo 2025-05-07 18:30:36 +08:00
parent b951310ca5
commit fa22c1a5a4
8 changed files with 28 additions and 11 deletions
Download patch file Download diff file Expand all files Collapse all files

7
koboldcpp.py
View file

@ -3681,7 +3681,10 @@ def RunServerMultiThreaded(addr, port, server_handler):
global exitcounter
exitcounter = 999
for i in range(numThreads):
threadArr[i].stop()
try:
threadArr[i].stop()
except Exception:
continue
sys.exit(0)
# Based on https://github.com/mathgeniuszach/xdialog/blob/main/xdialog/zenity_dialogs.py - MIT license | - Expanded version by Henk717
@ -6723,7 +6726,7 @@ if __name__ == '__main__':
compatgroup.add_argument("--usevulkan", help="Use Vulkan for GPU Acceleration. Can optionally specify one or more GPU Device ID (e.g. --usevulkan 0), leave blank to autodetect.", metavar=('[Device IDs]'), nargs='*', type=int, default=None)
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("--usecpu", help="Do not use any GPU acceleration (CPU Only)", action='store_true')
parser.add_argument("--contextsize", help="Controls the memory allocated for maximum context size, only change if you need more RAM for big contexts. (default 4096). Supported values are [256,512,1024,2048,3072,4096,6144,8192,10240,12288,14336,16384,20480,24576,28672,32768,40960,49152,57344,65536,81920,98304,114688,131072]. IF YOU USE ANYTHING ELSE YOU ARE ON YOUR OWN.",metavar=('[256,512,1024,2048,3072,4096,6144,8192,10240,12288,14336,16384,20480,24576,28672,32768,40960,49152,57344,65536,81920,98304,114688,131072]'), type=check_range(int,256,262144), default=4096)
parser.add_argument("--contextsize", help="Controls the memory allocated for maximum context size, only change if you need more RAM for big contexts. (default 4096).",metavar=('[256 to 262144]'), type=check_range(int,256,262144), default=4096)
parser.add_argument("--gpulayers", help="Set number of layers to offload to GPU when using GPU. Requires GPU. Set to -1 to try autodetect, set to 0 to disable GPU offload.",metavar=('[GPU layers]'), nargs='?', const=1, type=int, default=-1)
parser.add_argument("--tensor_split", help="For CUDA and Vulkan only, ratio to split tensors across multiple GPUs, space-separated list of proportions, e.g. 7 3", metavar=('[Ratios]'), type=float, nargs='+')