Update README.md

README.md

@@ -125,8 +125,8 @@ TODO
 | Benchmark                        |                |                               |
 |----------------------------------|----------------|-------------------------------|
 |                                  | Qwen3-32B      | Qwen3-32B-float8dq            | 
-| latency (batch_size=1)           | 9.1s           | TODO                          |
-| latency (batch_size=128)         | 12.45s         | TODO                          |
+| latency (batch_size=1)           | 9.1s           | 5.77s (-36.6%)                |
+| latency (batch_size=128)         | 12.45s         | 8.40s (-32.5%)                |
 | serving (num_prompts=1)          | TODO           | TODO                          |
 | serving (num_prompts=1000)       | TODO           | TODO                          |
 
@@ -156,7 +156,7 @@ wget https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered/r
 Server:
 ```Shell
 export MODEL=Qwen/Qwen3-32B # or pytorch/Qwen3-32B-float8dq
-VLLM_DISABLE_COMPILE_CACHE=1 vllm serve MODEL --tokenizer Qwen/Qwen3-32B -O3
+VLLM_DISABLE_COMPILE_CACHE=1 vllm serve $MODEL --tokenizer Qwen/Qwen3-32B -O3
 ```
 
 Client: