diff --git a/mlx/compile.cpp b/mlx/compile.cpp
index 9b52baa91..1f9556346 100644
--- a/mlx/compile.cpp
+++ b/mlx/compile.cpp
@@ -211,6 +211,8 @@ std::uintptr_t get_function_address(const std::function<T(U...)>& fun) {
 class CompilerCache {
  public:
   struct CacheEntry {
+    CacheEntry(Stream stream) : stream(stream) {};
+    Stream stream;
     std::vector<array> inputs;
     std::vector<array> outputs;
     std::vector<array> tape;
@@ -227,6 +229,7 @@ class CompilerCache {
       const std::vector<uint64_t>& constants) {
     // Find the cache entries for |fun_id|.
     std::vector<CacheEntry>& entries = cache_[fun_id];
+
     // Compare if 2 arrays have same shape and dtype.
     auto has_same_shape_and_dtype = [shapeless](
                                         const std::vector<array>& in1,
@@ -247,11 +250,16 @@ class CompilerCache {
       }
       return true;
     };
-    // Loop over entries and check inputs match i.e. shapes and types must be
-    // equal. Note this could get really slow if one compiles the same
-    // function with many different shapes. May want to store entries in a
-    // more easily searchable structure.
+    // Loop over entries and check:
+    // - Default stream and device match the entry's default stream
+    // - Inputs match i.e. shapes and types must be equal.
+    auto stream = default_stream(default_device());
     for (CacheEntry& entry : entries) {
+      // Check that the default stream and device match
+      if (entry.stream != stream) {
+        continue;
+      }
+
       // Check the inputs match and return if so
       if (has_same_shape_and_dtype(inputs, entry.inputs) &&
           constants == entry.constants) {
@@ -259,7 +267,7 @@ class CompilerCache {
       }
     }
     // Otherwise append a new cache entry
-    entries.push_back(CacheEntry{});
+    entries.push_back(CacheEntry{stream});
     return entries.back();
   }
 
diff --git a/python/src/array.cpp b/python/src/array.cpp
index c5a3c0cdd..e518f2765 100644
--- a/python/src/array.cpp
+++ b/python/src/array.cpp
@@ -184,10 +184,6 @@ void init_array(nb::module_& m) {
       R"pbdoc(
       A helper object to apply updates at specific indices.
       )pbdoc")
-      .def(
-          nb::init<const array&>(),
-          "x"_a,
-          nb::sig("def __init__(self, x: array)"))
       .def("__getitem__", &ArrayAt::set_indices, "indices"_a.none())
       .def("add", &ArrayAt::add, "value"_a)
       .def("subtract", &ArrayAt::subtract, "value"_a)
@@ -202,10 +198,6 @@ void init_array(nb::module_& m) {
       R"pbdoc(
       A helper object to iterate over the 1st dimension of an array.
       )pbdoc")
-      .def(
-          nb::init<const array&>(),
-          "x"_a,
-          nb::sig("def __init__(self, x: array)"))
       .def("__next__", &ArrayPythonIterator::next)
       .def("__iter__", [](const ArrayPythonIterator& it) { return it; });
 
diff --git a/python/src/stream.cpp b/python/src/stream.cpp
index 81260abb7..e1eb6b953 100644
--- a/python/src/stream.cpp
+++ b/python/src/stream.cpp
@@ -48,7 +48,6 @@ void init_stream(nb::module_& m) {
       R"pbdoc(
       A stream for running operations on a given device.
       )pbdoc")
-      .def(nb::init<int, Device>(), "index"_a, "device"_a)
       .def_ro("device", &Stream::device)
       .def(
           "__repr__",
diff --git a/tests/compile_tests.cpp b/tests/compile_tests.cpp
index d1ca59c99..a1559b7d3 100644
--- a/tests/compile_tests.cpp
+++ b/tests/compile_tests.cpp
@@ -719,3 +719,14 @@ TEST_CASE("test compile strides") {
     CHECK_EQ(out.strides().size(), 3);
   }
 }
+
+TEST_CASE("test compile change streams") {
+  auto cfun = compile(simple_fun);
+  auto out = cfun({array(1.0f), array(2.0f)})[0];
+  CHECK_EQ(out.primitive().stream(), default_stream(default_device()));
+
+  auto s = new_stream(default_device());
+  StreamContext sctx(s);
+  out = cfun({array(1.0f), array(2.0f)})[0];
+  CHECK_EQ(out.primitive().stream(), s);
+}