diff --git a/evalscope/evaluator/evaluator.py b/evalscope/evaluator/evaluator.py
index 1b78b2f..f894411 100644
--- a/evalscope/evaluator/evaluator.py
+++ b/evalscope/evaluator/evaluator.py
@@ -86,6 +86,7 @@ def __init__(self,
             **kwargs)
 
         # Get prompts from dataset
+        # TODO: support sampler
         self.prompts = self.data_adapter.gen_prompts(data_dict=self.dataset)
         del self.dataset
 
diff --git a/evalscope/perf/main.py b/evalscope/perf/main.py
index 969bae8..36b33e5 100644
--- a/evalscope/perf/main.py
+++ b/evalscope/perf/main.py
@@ -19,7 +19,9 @@ def run_perf_benchmark(args):
         args = Arguments(**args)
     elif isinstance(args, Namespace):
         args = Arguments.from_args(args)
-    seed_everything(args.seed)
+
+    if args.seed is not None:
+        seed_everything(args.seed)
 
     # Setup logger and output
     args.outputs_dir = get_output_path(args)