From c07526e46c98dd72c62822c8a5e0fb21e1aeeb16 Mon Sep 17 00:00:00 2001
From: zhyncs <me@zhyncs.com>
Date: Mon, 22 Jul 2024 18:23:33 +1000
Subject: [PATCH] fix: update bench serving (#694)

---
 python/sglang/bench_serving.py | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/python/sglang/bench_serving.py b/python/sglang/bench_serving.py
index e110be750..348307524 100644
--- a/python/sglang/bench_serving.py
+++ b/python/sglang/bench_serving.py
@@ -70,12 +70,10 @@ async def async_request_trt_llm(
     assert api_url.endswith("generate_stream")
 
     async with aiohttp.ClientSession(timeout=AIOHTTP_TIMEOUT) as session:
-        assert not request_func_input.use_beam_search
-        assert request_func_input.best_of == 1
         payload = {
             "accumulate_tokens": True,
             "text_input": request_func_input.prompt,
-            "temperature": 0.0,
+            "temperature": 0.000001,
             "top_p": 1.0,
             "max_tokens": request_func_input.output_len,
             "stream": True,