diff --git a/WORKSPACE b/WORKSPACE
index 8a119d8a236..44c02a0785c 100644
--- a/WORKSPACE
+++ b/WORKSPACE
@@ -50,7 +50,7 @@ new_local_repository(
 #    curl -L https://github.com/openxla/xla/archive/<git hash>.tar.gz | sha256sum
 #    and update the sha256 with the result.
 
-xla_hash = 'd28bfbdc366627c9ac9f57fcaa512ff04de19d6f'
+xla_hash = '8d06f3680ad046ea44f8e7159f52c728bb66c069'
 
 http_archive(
     name = "xla",
diff --git a/setup.py b/setup.py
index 1a808ec00aa..0e0f4971c05 100644
--- a/setup.py
+++ b/setup.py
@@ -65,7 +65,7 @@
 base_dir = os.path.dirname(os.path.abspath(__file__))
 
 USE_NIGHTLY = True  # whether to use nightly or stable libtpu and jax
-_date = '20250106'
+_date = '20250113'
 _libtpu_version = f'0.0.8'
 _jax_version = f'0.4.39'
 _jaxlib_version = f'0.4.39'
diff --git a/test/cpp/test_aten_xla_tensor_2.cpp b/test/cpp/test_aten_xla_tensor_2.cpp
index 92d3bce02ed..28b107b1e6f 100755
--- a/test/cpp/test_aten_xla_tensor_2.cpp
+++ b/test/cpp/test_aten_xla_tensor_2.cpp
@@ -513,8 +513,6 @@ TEST_F(AtenXlaTensorTest, TestLinalgVectorNormInDimsKeepDtype) {
 }
 
 TEST_F(AtenXlaTensorTest, TestLinalgEigh) {
-  // TODO: Broken by XLA pin update on 20250106.
-  GTEST_SKIP();
   // Hardcode the test input to avoid numerical instability from randomness,
   // which is a problem in eigenvalue decomposition.
   auto complex64 = [](float real, float imag) {
diff --git a/torch_xla/csrc/dl_convertor.cpp b/torch_xla/csrc/dl_convertor.cpp
index a2310f61d35..2f174a4af22 100644
--- a/torch_xla/csrc/dl_convertor.cpp
+++ b/torch_xla/csrc/dl_convertor.cpp
@@ -154,7 +154,7 @@ DLManagedTensor* toDLPack(const at::Tensor& input) {
 
   pack->shape = std::vector<int64_t>(pjrt_buffer->dimensions().begin(),
                                      pjrt_buffer->dimensions().end());
-  xla::Layout xla_layout = xla::GetXlaLayoutUnsafe(pjrt_buffer->layout());
+  xla::Layout xla_layout = pjrt_buffer->layout()->xla_layout();
   pack->strides = StridesForShape(pjrt_buffer->element_type(),
                                   pjrt_buffer->dimensions(), xla_layout);
   dt.shape = reinterpret_cast<std::int64_t*>(pack->shape.data());
diff --git a/torch_xla/csrc/runtime/BUILD b/torch_xla/csrc/runtime/BUILD
index 213761748c7..0e5b714cd1e 100644
--- a/torch_xla/csrc/runtime/BUILD
+++ b/torch_xla/csrc/runtime/BUILD
@@ -63,6 +63,7 @@ cc_library(
         "@xla//xla:literal_util",
         "@xla//xla/client:xla_computation",
         "@xla//xla/hlo/ir:hlo",
+        "@xla//xla/pjrt:pjrt_client",
     ],
 )
 
diff --git a/torch_xla/csrc/runtime/pjrt_computation_client.cc b/torch_xla/csrc/runtime/pjrt_computation_client.cc
index 30e648919b3..8caad6d230f 100644
--- a/torch_xla/csrc/runtime/pjrt_computation_client.cc
+++ b/torch_xla/csrc/runtime/pjrt_computation_client.cc
@@ -62,8 +62,7 @@ std::unordered_map<int, int> build_index_map(
 xla::Shape host_output_shape(xla::PjRtBuffer* buffer) {
   xla::Shape shape = xla::ShapeUtil::MakeShape(
       buffer->element_type(), buffer->logical_dimensions().value());
-  *shape.mutable_layout() = xla::GetXlaLayoutUnsafe(buffer->layout());
-
+  *shape.mutable_layout() = buffer->layout()->xla_layout();
   return xla::ShapeUtil::DeviceShapeToHostShape(shape);
 }