diff --git a/crates/uv-torch/src/backend.rs b/crates/uv-torch/src/backend.rs
index a60ff0cad..9db252c77 100644
--- a/crates/uv-torch/src/backend.rs
+++ b/crates/uv-torch/src/backend.rs
@@ -59,6 +59,8 @@ pub enum TorchMode {
     Auto,
     /// Use the CPU-only PyTorch index.
     Cpu,
+    /// Use the PyTorch index for CUDA 12.8.
+    Cu128,
     /// Use the PyTorch index for CUDA 12.6.
     Cu126,
     /// Use the PyTorch index for CUDA 12.5.
@@ -131,6 +133,7 @@ impl TorchStrategy {
                 }
             }
             TorchMode::Cpu => Ok(Self::Backend(TorchBackend::Cpu)),
+            TorchMode::Cu128 => Ok(Self::Backend(TorchBackend::Cu128)),
             TorchMode::Cu126 => Ok(Self::Backend(TorchBackend::Cu126)),
             TorchMode::Cu125 => Ok(Self::Backend(TorchBackend::Cu125)),
             TorchMode::Cu124 => Ok(Self::Backend(TorchBackend::Cu124)),
@@ -230,6 +233,7 @@ impl TorchStrategy {
 #[derive(Debug, Clone, Eq, PartialEq)]
 pub enum TorchBackend {
     Cpu,
+    Cu128,
     Cu126,
     Cu125,
     Cu124,
@@ -260,6 +264,7 @@ impl TorchBackend {
     fn index_url(&self) -> &'static IndexUrl {
         match self {
             Self::Cpu => &CPU_INDEX_URL,
+            Self::Cu128 => &CU128_INDEX_URL,
             Self::Cu126 => &CU126_INDEX_URL,
             Self::Cu125 => &CU125_INDEX_URL,
             Self::Cu124 => &CU124_INDEX_URL,
@@ -290,10 +295,11 @@ impl TorchBackend {
 /// Linux CUDA driver versions and the corresponding CUDA versions.
 ///
 /// See: <https://github.com/pmeier/light-the-torch/blob/33397cbe45d07b51ad8ee76b004571a4c236e37f/light_the_torch/_cb.py#L150-L213>
-static LINUX_DRIVERS: LazyLock<[(TorchBackend, Version); 23]> = LazyLock::new(|| {
+static LINUX_DRIVERS: LazyLock<[(TorchBackend, Version); 24]> = LazyLock::new(|| {
     [
         // Table 2 from
         // https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html
+        (TorchBackend::Cu128, Version::new([525, 60, 13])),
         (TorchBackend::Cu126, Version::new([525, 60, 13])),
         (TorchBackend::Cu125, Version::new([525, 60, 13])),
         (TorchBackend::Cu124, Version::new([525, 60, 13])),
@@ -327,10 +333,11 @@ static LINUX_DRIVERS: LazyLock<[(TorchBackend, Version); 23]> = LazyLock::new(||
 /// Windows CUDA driver versions and the corresponding CUDA versions.
 ///
 /// See: <https://github.com/pmeier/light-the-torch/blob/33397cbe45d07b51ad8ee76b004571a4c236e37f/light_the_torch/_cb.py#L150-L213>
-static WINDOWS_CUDA_VERSIONS: LazyLock<[(TorchBackend, Version); 23]> = LazyLock::new(|| {
+static WINDOWS_CUDA_VERSIONS: LazyLock<[(TorchBackend, Version); 24]> = LazyLock::new(|| {
     [
         // Table 2 from
         // https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html
+        (TorchBackend::Cu128, Version::new([528, 33])),
         (TorchBackend::Cu126, Version::new([528, 33])),
         (TorchBackend::Cu125, Version::new([528, 33])),
         (TorchBackend::Cu124, Version::new([528, 33])),
@@ -363,6 +370,8 @@ static WINDOWS_CUDA_VERSIONS: LazyLock<[(TorchBackend, Version); 23]> = LazyLock
 
 static CPU_INDEX_URL: LazyLock<IndexUrl> =
     LazyLock::new(|| IndexUrl::from_str("https://download.pytorch.org/whl/cpu").unwrap());
+static CU128_INDEX_URL: LazyLock<IndexUrl> =
+    LazyLock::new(|| IndexUrl::from_str("https://download.pytorch.org/whl/cu128").unwrap());
 static CU126_INDEX_URL: LazyLock<IndexUrl> =
     LazyLock::new(|| IndexUrl::from_str("https://download.pytorch.org/whl/cu126").unwrap());
 static CU125_INDEX_URL: LazyLock<IndexUrl> =
diff --git a/docs/reference/cli.md b/docs/reference/cli.md
index 8851ca2c0..f8399bcd2 100644
--- a/docs/reference/cli.md
+++ b/docs/reference/cli.md
@@ -6073,6 +6073,8 @@ uv pip compile [OPTIONS] <SRC_FILE|--group <GROUP>>
 
 <li><code>cpu</code>:  Use the CPU-only PyTorch index</li>
 
+<li><code>cu128</code>:  Use the PyTorch index for CUDA 12.8</li>
+
 <li><code>cu126</code>:  Use the PyTorch index for CUDA 12.6</li>
 
 <li><code>cu125</code>:  Use the PyTorch index for CUDA 12.5</li>
@@ -6548,6 +6550,8 @@ uv pip sync [OPTIONS] <SRC_FILE>...
 
 <li><code>cpu</code>:  Use the CPU-only PyTorch index</li>
 
+<li><code>cu128</code>:  Use the PyTorch index for CUDA 12.8</li>
+
 <li><code>cu126</code>:  Use the PyTorch index for CUDA 12.6</li>
 
 <li><code>cu125</code>:  Use the PyTorch index for CUDA 12.5</li>
@@ -7095,6 +7099,8 @@ uv pip install [OPTIONS] <PACKAGE|--requirements <REQUIREMENTS>|--editable <EDIT
 
 <li><code>cpu</code>:  Use the CPU-only PyTorch index</li>
 
+<li><code>cu128</code>:  Use the PyTorch index for CUDA 12.8</li>
+
 <li><code>cu126</code>:  Use the PyTorch index for CUDA 12.6</li>
 
 <li><code>cu125</code>:  Use the PyTorch index for CUDA 12.5</li>
diff --git a/uv.schema.json b/uv.schema.json
index edaf3c9c4..aab9c9e45 100644
--- a/uv.schema.json
+++ b/uv.schema.json
@@ -2321,6 +2321,13 @@
             "cpu"
           ]
         },
+        {
+          "description": "Use the PyTorch index for CUDA 12.8.",
+          "type": "string",
+          "enum": [
+            "cu128"
+          ]
+        },
         {
           "description": "Use the PyTorch index for CUDA 12.6.",
           "type": "string",