Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-18 16:55:14 +02:00
Code Issues Packages Projects Releases Wiki Activity

bump default gen amount by 128 to 896

Browse Source
This commit is contained in:
Concedo
2025-12-14 22:17:31 +08:00
parent e46a6a2796
commit bca0258c2a
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
koboldcpp.py
+3 -3
View File
@@ -5189,7 +5189,7 @@ def show_gui():
jinja_tools_var = ctk.IntVar(value=0)
moeexperts_var = ctk.StringVar(value=str(-1))
moecpu_var = ctk.StringVar(value=str(0))
defaultgenamt_var = ctk.StringVar(value=str(768))
defaultgenamt_var = ctk.StringVar(value=str(896))
genlimit_var = ctk.StringVar(value=str(0))
nobostoken_var = ctk.IntVar(value=0)
override_kv_var = ctk.StringVar(value="")
@@ -6249,7 +6249,7 @@ def show_gui():
args.overridenativecontext = 0
args.moeexperts = int(moeexperts_var.get()) if moeexperts_var.get()!="" else -1
args.moecpu = int(moecpu_var.get()) if moecpu_var.get()!="" else 0
args.defaultgenamt = int(defaultgenamt_var.get()) if defaultgenamt_var.get()!="" else 768
args.defaultgenamt = int(defaultgenamt_var.get()) if defaultgenamt_var.get()!="" else 896
args.genlimit = int(genlimit_var.get()) if genlimit_var.get()!="" else 0
args.nobostoken = (nobostoken_var.get()==1)
args.jinja = (jinja_var.get()==1)
@@ -8409,7 +8409,7 @@ if __name__ == '__main__':
advparser.add_argument("--nomodel", help="Allows you to launch the GUI alone, without selecting any model.", action='store_true')
advparser.add_argument("--moeexperts", metavar=('[num of experts]'), help="How many experts to use for MoE models (default=follow gguf)", type=int, default=-1)
advparser.add_argument("--moecpu","--n-cpu-moe", "-ncmoe", metavar=('[layers affected]'), help="Keep the Mixture of Experts (MoE) weights of the first N layers in the CPU. If no value is provided, applies to all layers.", nargs='?', const=999, type=int, default=0)
advparser.add_argument("--defaultgenamt", help="How many tokens to generate by default, if not specified. Must be smaller than context size. Usually, your frontend GUI will override this.", type=check_range(int,64,8192), default=768)
advparser.add_argument("--defaultgenamt", help="How many tokens to generate by default, if not specified. Must be smaller than context size. Usually, your frontend GUI will override this.", type=check_range(int,64,8192), default=896)
advparser.add_argument("--nobostoken", help="Prevents BOS token from being added at the start of any prompt. Usually NOT recommended for most models.", action='store_true')
advparser.add_argument("--enableguidance", help="Enables the use of Classifier-Free-Guidance, which allows the use of negative prompts. Has performance and memory impact.", action='store_true')
advparser.add_argument("--maxrequestsize", metavar=('[size in MB]'), help="Specify a max request payload size. Any requests to the server larger than this size will be dropped. Do not change if unsure.", type=int, default=32)