diff --git a/README.md b/README.md
index 8266d0b2..621fc91c 100644
--- a/README.md
+++ b/README.md
@@ -23,12 +23,6 @@ set the CFG scale to something like 3.5-7
 forge doesn't seem to work with all quantized model, Q4_K_S fail, but Q4_1 work  
 refer to screenshot for working settings.  
 
-To update to torch 2.7.0 with cuda 12.8 on windows and install sage attention, navigate to your root directory after initial installation ie c:/chromaforge/ and run these commands:  
-venv/scripts/activate  
-pip install torch==2.7.0+cu128 torchvision==0.22.0+cu128 --index-url https://download.pytorch.org/whl/cu128  
-pip install -U "triton-windows<3.4"  
-pip install .\sageattention-2.1.1+cu128torch2.7.0-cp310-cp310-win_amd64.whl  
-
 there's a bunch of extra samplers/schedulers at these places:  
 https://github.com/DenOfEquity/webUI_ExtraSchedulers
 https://github.com/MisterChief95/sd-forge-extra-samplers
@@ -51,7 +45,8 @@ Donate to Lodestone (training is bookoo expensive and crowdfunded): https://ko-f
 Discord: http://discord.gg/SQVcWVbqKx  
 
 ## Changlog
-
+10/15/2025  
+    Upgrade default torch version to 2.7.1 to support 50xx series gpu.  
 9/29/2025  
     Implement diffusers style T5 embedding mask as seen here: https://github.com/huggingface/diffusers/pull/12263 seems to help alot with details.  
 9/26/2025  
diff --git a/modules/launch_utils.py b/modules/launch_utils.py
index 420d9b19..59ff5e1b 100644
--- a/modules/launch_utils.py
+++ b/modules/launch_utils.py
@@ -361,8 +361,8 @@ def requirements_met(requirements_file):
 
 
 def prepare_environment():
-    torch_index_url = os.environ.get('TORCH_INDEX_URL', "https://download.pytorch.org/whl/cu121")
-    torch_command = os.environ.get('TORCH_COMMAND', f"pip install torch==2.3.1 torchvision==0.18.1 --extra-index-url {torch_index_url}")
+    torch_index_url = os.environ.get('TORCH_INDEX_URL', "https://download.pytorch.org/whl/cu128")
+    torch_command = os.environ.get('TORCH_COMMAND', f"pip install torch==2.7.1+cu128 torchvision --index-url https://download.pytorch.org/whl/cu128")
     if args.use_ipex:
         if platform.system() == "Windows":
             # The "Nuullll/intel-extension-for-pytorch" wheels were built from IPEX source for Intel Arc GPU: https://github.com/intel/intel-extension-for-pytorch/tree/xpu-main
diff --git a/requirements_versions.txt b/requirements_versions.txt
index a6026209..bca09346 100644
--- a/requirements_versions.txt
+++ b/requirements_versions.txt
@@ -27,7 +27,7 @@ scikit-image==0.21.0
 spandrel==0.3.4
 spandrel-extra-arches==0.1.1
 tomesd==0.1.3
-torch
+torch==2.7.1
 torchdiffeq==0.2.3
 torchsde==0.2.6
 transformers==4.46.1
diff --git a/sageattention-2.1.1+cu128torch2.7.0-cp310-cp310-win_amd64.whl b/sageattention-2.1.1+cu128torch2.7.0-cp310-cp310-win_amd64.whl
deleted file mode 100644
index 91e5dd29..00000000
Binary files a/sageattention-2.1.1+cu128torch2.7.0-cp310-cp310-win_amd64.whl and /dev/null differ