tanner/note-committer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Keep model loaded in memory, saves 5 seconds

Browse Source
This commit is contained in:
Tanner Collin 2025-01-20 18:24:59 +00:00
parent 587bc67416
commit dcd8af2895
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
main.py
View File

@ -23,7 +23,7 @@ def controller_message(message):
return False return False
def llama(prompt): def llama(prompt):
data = dict(model='llama3.1', prompt=prompt, stream=False) data = dict(model='llama3.1', prompt=prompt, stream=False, keep_alive=-1)
try: try:
r = requests.post(LLAMA_URL, json=data, timeout=20) r = requests.post(LLAMA_URL, json=data, timeout=20)
r.raise_for_status() r.raise_for_status()