build: Update Windows torch versions (NVIDIA#309)

Author: sestephens-nv

requirements-dev-windows.txt

@@ -1,9 +1,9 @@
---pre --extra-index-url https://download.pytorch.org/whl/nightly/cu121
-# torch is CPU-only on Windows, so need to specify a torch version with GPU support
-torch==2.1.0.dev20230828+cu121
-torchvision==0.16.0.dev20230828
-torchdata==0.7.0.dev20230828
-torchtext==0.16.0.dev20230828
+--extra-index-url https://download.pytorch.org/whl/cu121
+# Default torch is CPU-only on Windows, so need to specify a torch version with GPU support
+torch==2.1.0+cu121
+torchvision==0.16.0+cu121
+torchdata==0.7.0
+torchtext==0.16.0+cpu
 tokenizers==0.13.3
 transformers==4.31.0
 diffusers==0.15.0

requirements-windows.txt

@@ -1,10 +1,10 @@
 build
---pre --extra-index-url https://download.pytorch.org/whl/nightly/cu121
-# torch is CPU-only on Windows, so need to specify a torch version with GPU support
-torch==2.1.0.dev20230828+cu121
-torchvision==0.16.0.dev20230828
-torchdata==0.7.0.dev20230828
-torchtext==0.16.0.dev20230828
+--extra-index-url https://download.pytorch.org/whl/cu121
+# Default torch is CPU-only on Windows, so need to specify a torch version with GPU support
+torch==2.1.0+cu121
+torchvision==0.16.0+cu121
+torchdata==0.7.0
+torchtext==0.16.0+cpu
 tokenizers==0.13.3
 transformers==4.31.0
 diffusers==0.15.0

windows/README.md

@@ -30,7 +30,7 @@ Prerequisites:
 - [TensorRT 9.1.0.4 for TensorRT-LLM](https://developer.nvidia.com/downloads/compute/machine-learning/tensorrt/secure/9.1.0/tars/tensorrt-9.1.0.4.windows10.x86_64.cuda-12.2.llm.beta.zip)
 
 ```
-pip install tensorrt_llm --extra-index-url https://pypi.nvidia.com --extra-index-url https://download.pytorch.org/whl/nightly/cu121
+pip install tensorrt_llm --extra-index-url https://pypi.nvidia.com --extra-index-url https://download.pytorch.org/whl/cu121
 ```
 
 ## Detailed Setup
@@ -118,7 +118,7 @@ The above command will generate `build\tensorrt_llm-0.5.0-py3-none-any.whl`. Oth
 
 To download and install the wheel, in Powershell, run:
 ```
-pip install tensorrt_llm --extra-index-url https://pypi.nvidia.com --extra-index-url https://download.pytorch.org/whl/nightly/cu121
+pip install tensorrt_llm --extra-index-url https://pypi.nvidia.com --extra-index-url https://download.pytorch.org/whl/cu121
 ```
 
 Alternatively, if you built the wheel from source, run: