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

Merge branch 'master' into concedo_experimental

Browse source 
# Conflicts:
#	.github/workflows/build.yml
#	.gitignore
#	CMakeLists.txt
#	Makefile
#	Package.swift
#	README.md
#	ggml-cuda.cu
#	llama.cpp
#	llama.h
#	scripts/sync-ggml.sh
#	tests/CMakeLists.txt
This commit is contained in:
Concedo 2023-12-08 17:42:26 +08:00
commit ec21fa7712
34 changed files with 5887 additions and 1435 deletions
Download patch file Download diff file Expand all files Collapse all files

2
koboldcpp.py
View file

@ -24,7 +24,6 @@ class load_model_inputs(ctypes.Structure):
("blasthreads", ctypes.c_int),
("max_context_length", ctypes.c_int),
("batch_size", ctypes.c_int),
("f16_kv", ctypes.c_bool),
("low_vram", ctypes.c_bool),
("use_mmq", ctypes.c_bool),
("executable_path", ctypes.c_char_p),
@ -235,7 +234,6 @@ def load_model(model_filename):
inputs.low_vram = (True if (args.usecublas and "lowvram" in args.usecublas) else False)
inputs.use_mmq = (True if (args.usecublas and "mmq" in args.usecublas) else False)
inputs.blasthreads = args.blasthreads
inputs.f16_kv = True
inputs.use_mmap = (not args.nommap)
inputs.use_mlock = args.usemlock
inputs.lora_filename = "".encode("UTF-8")