mirror of
https://github.com/LostRuins/koboldcpp.git
synced 2025-09-10 17:14:36 +00:00
bump token limits
This commit is contained in:
parent
8bf6f7f8b0
commit
eb86cd4027
2 changed files with 5 additions and 5 deletions
|
@ -11,9 +11,9 @@ import argparse
|
|||
import json, sys, http.server, time, asyncio, socket, threading
|
||||
from concurrent.futures import ThreadPoolExecutor
|
||||
|
||||
stop_token_max = 10
|
||||
sampler_order_max = 7
|
||||
ban_token_max = 10
|
||||
stop_token_max = 16
|
||||
ban_token_max = 16
|
||||
tensor_split_max = 16
|
||||
|
||||
class load_model_inputs(ctypes.Structure):
|
||||
|
@ -749,7 +749,7 @@ def RunServerMultiThreaded(addr, port, embedded_kailite = None):
|
|||
exitcounter = 999
|
||||
self.httpd.server_close()
|
||||
|
||||
numThreads = 10
|
||||
numThreads = 12
|
||||
threadArr = []
|
||||
for i in range(numThreads):
|
||||
threadArr.append(Thread(i))
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue