[Benchmark] Don't default to `temperature==0` in `vllm bench serve` (#32723)

Signed-off-by: Nick Hill <nickhill123@gmail.com>

vllm/benchmarks/lib/endpoint_request_func.py

@@ -160,7 +160,6 @@ async def async_request_openai_completions(
         if request_func_input.model_name
         else request_func_input.model,
         "prompt": request_func_input.prompt,
-        "temperature": 0.0,
         "repetition_penalty": 1.0,
         "max_tokens": request_func_input.output_len,
         "logprobs": request_func_input.logprobs,
@@ -294,7 +293,6 @@ async def async_request_openai_chat_completions(
         "messages": [
             {"role": "user", "content": content},
         ],
-        "temperature": 0.0,
         "max_completion_tokens": request_func_input.output_len,
         "stream": True,
         "stream_options": {
@@ -389,7 +387,6 @@ async def async_request_openai_audio(
         "model": request_func_input.model_name
         if request_func_input.model_name
         else request_func_input.model,
-        "temperature": 0.0,
         "max_completion_tokens": request_func_input.output_len,
         "stream": True,
         "language": "en",

vllm/benchmarks/serve.py

@@ -1419,8 +1419,7 @@ def add_cli_args(parser: argparse.ArgumentParser):
         type=float,
         default=None,
         help="Temperature sampling parameter. Only has effect on "
-        "openai-compatible backends. If not specified, default to greedy "
-        "decoding (i.e. temperature==0.0).",
+        "openai-compatible backends.",
     )
     sampling_group.add_argument(
         "--frequency-penalty",
@@ -1634,7 +1633,12 @@ async def main_async(args: argparse.Namespace) -> dict[str, Any]:
             )
 
         if "temperature" not in sampling_params:
-            sampling_params["temperature"] = 0.0  # Default to greedy decoding.
+            print(
+                "WARNING: vllm bench serve no longer sets temperature==0 (greedy) "
+                "in requests by default. The default will be determined on the "
+                "server side and can be model/API specific. "
+                "For the old behavior, include --temperature=0."
+            )
 
         default_percentile_metrics = "ttft,tpot,itl"
     else: