Enable cuda graph by default (#612)

2024-07-13 05:29:46 -07:00
parent 396a69240f
commit 665815969a
10 changed files with 331 additions and 84 deletions
--- a/python/sglang/bench_latency.py
+++ b/python/sglang/bench_latency.py
@@ -30,7 +30,6 @@ import argparse
 import dataclasses
 import logging
 import multiprocessing
-import os
 import time