vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-10 09:04:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

added support for OAI chat completions adapter file, added default stop sequences to prevent chat compl leakage

Browse source
This commit is contained in:
Concedo 2024-04-07 10:35:20 +08:00
parent 0061299cce
commit 6166fdfde4
2 changed files with 28 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

3
class.py
View file

@ -273,7 +273,8 @@ class model_backend(InferenceModel):
unbantokens=False, bantokens=None, usemirostat=None, forceversion=0, nommap=self.kcpp_nommap,
usemlock=False, noavx2=self.kcpp_noavx2, debugmode=self.kcpp_debugmode, skiplauncher=True, hordeconfig=None, noblas=self.kcpp_noblas,
useclblast=self.kcpp_useclblast, usecublas=self.kcpp_usecublas, usevulkan=self.kcpp_usevulkan, gpulayers=self.kcpp_gpulayers, tensor_split=self.kcpp_tensor_split, config=None,
onready='', multiuser=False, foreground=False, preloadstory=None, noshift=False, remotetunnel=False, ssl=False, benchmark=None, nocertify=False, sdconfig=None, mmproj=None, password=None)
onready='', multiuser=False, foreground=False, preloadstory=None, noshift=False, remotetunnel=False, ssl=False, benchmark=None, nocertify=False, sdconfig=None, mmproj=None,
password=None, chatcompletionsadapter=None)
#koboldcpp.main(kcppargs,False) #initialize library without enabling Lite http server