Update README.md

README.md

@@ -26,8 +26,7 @@ from INT4 checkpoint from 36.98 to 43.13, while bfloat16 baseline accuracy is 46
 # Inference with vLLM
 Install vllm nightly and torchao nightly to get some recent changes:
 ```
-pip install vllm --pre --extra-index-url https://wheels.vllm.ai/nightly
-pip install --pre torchao --index-url https://download.pytorch.org/whl/nightly/cu126
+pip install --pre torchao torch vllm --index-url https://download.pytorch.org/whl/nightly/cu128
 ```
 
 ## Serving
@@ -125,9 +124,8 @@ Install the required packages:
 ```Shell
 # for compatibility with modeling file in checkpoint
 pip install transformers==4.53.0
-pip install --pre torchao --index-url https://download.pytorch.org/whl/nightly/cu126
-pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126
 pip install accelerate
+pip install --pre torchao torch --index-url https://download.pytorch.org/whl/nightly/cu128
 ```