diff --git a/.github/workflows/publish.yml b/.github/workflows/publish.yml
index 4b9fc3d04d872..d79681f03b003 100644
--- a/.github/workflows/publish.yml
+++ b/.github/workflows/publish.yml
@@ -49,7 +49,7 @@ jobs:
       matrix:
           os: ['ubuntu-20.04']
           python-version: ['3.8', '3.9', '3.10', '3.11']
-          pytorch-version: ['2.2.1']  # Must be the most recent version that meets requirements-cuda.txt.
+          pytorch-version: ['2.3.0']  # Must be the most recent version that meets requirements-cuda.txt.
           cuda-version: ['11.8', '12.1']
 
     steps:
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 1558dbf313ce7..f817f3382c5e1 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -31,7 +31,7 @@ set(HIP_SUPPORTED_ARCHS "gfx906;gfx908;gfx90a;gfx940;gfx941;gfx942;gfx1030;gfx11
 # requirements.txt files and should be kept consistent.  The ROCm torch
 # versions are derived from Dockerfile.rocm
 #
-set(TORCH_SUPPORTED_VERSION_CUDA "2.2.1")
+set(TORCH_SUPPORTED_VERSION_CUDA "2.3.0")
 set(TORCH_SUPPORTED_VERSION_ROCM_5X "2.0.1")
 set(TORCH_SUPPORTED_VERSION_ROCM_6X "2.1.1")
 
diff --git a/Dockerfile b/Dockerfile
index d1d29177b0f44..e471a6e93b963 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -85,7 +85,7 @@ FROM dev as flash-attn-builder
 ARG max_jobs=2
 ENV MAX_JOBS=${max_jobs}
 # flash attention version
-ARG flash_attn_version=v2.5.6
+ARG flash_attn_version=v2.5.8
 ENV FLASH_ATTN_VERSION=${flash_attn_version}
 
 WORKDIR /usr/src/flash-attention-v2
diff --git a/pyproject.toml b/pyproject.toml
index 2e026c1ac8911..6a448defc16e1 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -5,7 +5,7 @@ requires = [
     "ninja",
     "packaging",
     "setuptools >= 49.4.0",
-    "torch == 2.2.1",
+    "torch == 2.3.0",
     "wheel",
 ]
 build-backend = "setuptools.build_meta"
diff --git a/requirements-build.txt b/requirements-build.txt
index 2bc07fb152aac..1a07a94e82e04 100644
--- a/requirements-build.txt
+++ b/requirements-build.txt
@@ -3,5 +3,5 @@ cmake>=3.21
 ninja
 packaging
 setuptools>=49.4.0
-torch==2.2.1
+torch==2.3.0
 wheel
diff --git a/requirements-cpu.txt b/requirements-cpu.txt
index e911ad03295f0..b739642d8d344 100644
--- a/requirements-cpu.txt
+++ b/requirements-cpu.txt
@@ -2,5 +2,5 @@
 -r requirements-common.txt
 
 # Dependencies for x86_64 CPUs
-torch == 2.2.1+cpu
+torch == 2.3.0+cpu
 triton >= 2.2.0  # FIXME(woosuk): This is a hack to avoid import error.
\ No newline at end of file
diff --git a/requirements-cuda.txt b/requirements-cuda.txt
index 1bddae4c6f40f..6548d7a6684b2 100644
--- a/requirements-cuda.txt
+++ b/requirements-cuda.txt
@@ -5,5 +5,5 @@
 ray >= 2.9
 nvidia-ml-py # for pynvml package
 vllm-nccl-cu12>=2.18,<2.19  # for downloading nccl library
-torch == 2.2.1
-xformers == 0.0.25  # Requires PyTorch 2.2.1
+torch == 2.3.0
+xformers == 0.0.26.post1  # Requires PyTorch 2.3.0