GLM-5.3-Flash

Running on Zero

apathy-exe commited on Sep 10

Commit

40fbb2c

verified ·

1 Parent(s): 1c7e69a

Update app.py

Files changed (1) hide show

app.py CHANGED Viewed

@@ -112,7 +112,7 @@ def start_llama_server():
         "-tb", "16",
         "-fa", "on",
         "--parallel", "1",
-        "--load-mode", "dio",
         #"--spec-type", "draft-mtp",
         #"--spec-draft-n-max", "4",
         "--cache-type-k", "q4_0",

         "-tb", "16",
         "-fa", "on",
         "--parallel", "1",
+        "--load-mode", "mmap",
         #"--spec-type", "draft-mtp",
         #"--spec-draft-n-max", "4",
         "--cache-type-k", "q4_0",