Fix cuda graph with flashinfer (#675)

2024-07-20 02:43:55 -07:00
parent 06487f126e
commit 490a1f39dd
2 changed files with 3 additions and 3 deletions
--- a/benchmark/gsm8k/bench_sglang.py
+++ b/benchmark/gsm8k/bench_sglang.py
@@ -64,7 +64,7 @@ def main(args):
    @sgl.function
    def few_shot_gsm8k(s, question):
        s += few_shot_examples + question
-        s += sgl.gen("answer", max_tokens=256, stop="Question")
+        s += sgl.gen("answer", max_tokens=512, stop="Question")

    #####################################
    ########## SGL Program End ##########