diff --git a/vllm/entrypoints/cli/benchmark/base.py b/vllm/entrypoints/cli/benchmark/base.py
index c41b2c586781..94fb415f581f 100644
--- a/vllm/entrypoints/cli/benchmark/base.py
+++ b/vllm/entrypoints/cli/benchmark/base.py
@@ -32,6 +32,7 @@ class BenchmarkSubcommandBase(CLISubcommand):
         parser = subparsers.add_parser(
             self.name,
             help=self.help,
+            description=self.help,
             usage=f"vllm bench {self.name} [options]")
         self.add_cli_args(parser)
         return parser
diff --git a/vllm/entrypoints/cli/benchmark/main.py b/vllm/entrypoints/cli/benchmark/main.py
index 7583540920d3..1bcb25be2fca 100644
--- a/vllm/entrypoints/cli/benchmark/main.py
+++ b/vllm/entrypoints/cli/benchmark/main.py
@@ -33,6 +33,7 @@ class BenchmarkSubcommand(CLISubcommand):
         bench_parser = subparsers.add_parser(
             "bench",
             help="vLLM bench subcommand.",
+            description="vLLM bench subcommand.",
             usage="vllm bench <bench_type> [options]")
         bench_subparsers = bench_parser.add_subparsers(required=True,
                                                        dest="bench_type")
diff --git a/vllm/entrypoints/cli/openai.py b/vllm/entrypoints/cli/openai.py
index 21a7d48b75c1..1d1bba1d49ce 100644
--- a/vllm/entrypoints/cli/openai.py
+++ b/vllm/entrypoints/cli/openai.py
@@ -126,7 +126,8 @@ class ChatCommand(CLISubcommand):
             subparsers: argparse._SubParsersAction) -> FlexibleArgumentParser:
         chat_parser = subparsers.add_parser(
             "chat",
-            help="Generate chat completions via the running API server",
+            help="Generate chat completions via the running API server.",
+            description="Generate chat completions via the running API server.",
             usage="vllm chat [options]")
         _add_query_options(chat_parser)
         chat_parser.add_argument(
@@ -162,7 +163,9 @@ class CompleteCommand(CLISubcommand):
         complete_parser = subparsers.add_parser(
             "complete",
             help=("Generate text completions based on the given prompt "
-                  "via the running API server"),
+                  "via the running API server."),
+            description=("Generate text completions based on the given prompt "
+                         "via the running API server."),
             usage="vllm complete [options]")
         _add_query_options(complete_parser)
         return complete_parser
diff --git a/vllm/entrypoints/cli/serve.py b/vllm/entrypoints/cli/serve.py
index e89ac4e21999..5c8781b50d2c 100644
--- a/vllm/entrypoints/cli/serve.py
+++ b/vllm/entrypoints/cli/serve.py
@@ -34,7 +34,8 @@ class ServeSubcommand(CLISubcommand):
             subparsers: argparse._SubParsersAction) -> FlexibleArgumentParser:
         serve_parser = subparsers.add_parser(
             "serve",
-            help="Start the vLLM OpenAI Compatible API server",
+            help="Start the vLLM OpenAI Compatible API server.",
+            description="Start the vLLM OpenAI Compatible API server.",
             usage="vllm serve [model_tag] [options]")
         serve_parser.add_argument("model_tag",
                                   type=str,