diff --git a/experimental/torch_xla2/test/test_ops.py b/experimental/torch_xla2/test/test_ops.py
index 3396eec2c80..3ace00ea495 100644
--- a/experimental/torch_xla2/test/test_ops.py
+++ b/experimental/torch_xla2/test/test_ops.py
@@ -72,7 +72,6 @@
     "special.scaled_modified_bessel_k1",
     "special.spherical_bessel_j0",
     "special.zeta",
-    "svd_lowrank",
     "unfold_copy",
     "unfold",
     "randint",
@@ -122,6 +121,7 @@
              "linalg.pinv": (8e-1, 2e0),
              "linalg.svd": (1e0, 1e0),
              "svd": (1e0, 1e0),
+             "svd_lowrank": (1e0, 1e0),
              "matrix_exp": (2e-1, 2e-4),
              "cdist": (5e1, 3e0)}
 
diff --git a/experimental/torch_xla2/torch_xla2/ops/jaten.py b/experimental/torch_xla2/torch_xla2/ops/jaten.py
index 6160a61fb6d..9b24b075cec 100644
--- a/experimental/torch_xla2/torch_xla2/ops/jaten.py
+++ b/experimental/torch_xla2/torch_xla2/ops/jaten.py
@@ -4559,8 +4559,8 @@ def _aten__linalg_slogdet(input):
 
 # torch.linalg.svd
 @op(torch.ops.aten._linalg_svd)
-def _aten__linalg_svd(a, full_matrices=True):
-  return jnp.linalg.svd(a, full_matrices=full_matrices)
+def _aten__linalg_svd(a, full_matrices=False, **kwargs):
+  return jnp.linalg.svd(a, full_matrices=full_matrices, **kwargs)
 
 
 # torch.linalg.pinv
diff --git a/experimental/torch_xla2/torch_xla2/ops/jtorch.py b/experimental/torch_xla2/torch_xla2/ops/jtorch.py
index c8b49c35c24..110fa05dbf7 100644
--- a/experimental/torch_xla2/torch_xla2/ops/jtorch.py
+++ b/experimental/torch_xla2/torch_xla2/ops/jtorch.py
@@ -355,8 +355,8 @@ def linalg_solve_ex(a, b):
   return res, info
 
 @register_function(torch.linalg.svd)
-def linalg_svd(a, full_matrices=True, **kwargs):
-  return jaten._aten__linalg_svd(a, full_matrices=full_matrices, **kwargs)
+def linalg_svd(a, full_matrices=True):
+  return jaten._aten__linalg_svd(a, full_matrices=full_matrices)
 
 @register_function(torch.linalg.matrix_power)
 def matrix_power(A, n, *, out=None):