[Misc] Fix the benchmark's README and improve the error messages for the...

[Misc] Fix the benchmark's README and improve the error messages for the benchmark's argument checks (#22654)
Signed-off-by: tanruixiang <tanruixiang0104@gmail.com>

benchmarks/README.md

@@ -194,6 +194,7 @@ vllm serve Qwen/Qwen2-VL-7B-Instruct
 ```bash
 vllm bench serve \
   --backend openai-chat \
+  --endpoint-type openai-chat \
   --model Qwen/Qwen2-VL-7B-Instruct \
   --endpoint /v1/chat/completions \
   --dataset-name hf \
@@ -230,6 +231,7 @@ vllm serve Qwen/Qwen2-VL-7B-Instruct
 ```bash
 vllm bench serve \
   --backend openai-chat \
+  --endpoint-type openai-chat \  
   --model Qwen/Qwen2-VL-7B-Instruct \
   --endpoint /v1/chat/completions \
   --dataset-name hf \
@@ -244,6 +246,7 @@ vllm bench serve \
 ```bash
 vllm bench serve \
   --backend openai-chat \
+  --endpoint-type openai-chat \  
   --model Qwen/Qwen2-VL-7B-Instruct \
   --endpoint /v1/chat/completions \
   --dataset-name hf \

vllm/benchmarks/datasets.py

@@ -740,10 +740,11 @@ def get_samples(args, tokenizer) -> list[SampleRequest]:
                 "openai-chat",
                 "openai-audio",
         ]:
-            # multi-modal benchmark is only available on OpenAI Chat backend.
+            # multi-modal benchmark is only available on OpenAI Chat
+            # endpoint-type.
             raise ValueError(
                 "Multi-modal content is only supported on 'openai-chat' and "
-                "'openai-audio' backend.")
+                "'openai-audio' endpoint-type.")
         input_requests = dataset_class(
             dataset_path=args.dataset_path,
             dataset_subset=args.hf_subset,