diff --git a/include/cudaq/Optimizer/Dialect/CC/CCTypes.td b/include/cudaq/Optimizer/Dialect/CC/CCTypes.td
index f85a8be11c..d8a5820abe 100644
--- a/include/cudaq/Optimizer/Dialect/CC/CCTypes.td
+++ b/include/cudaq/Optimizer/Dialect/CC/CCTypes.td
@@ -195,17 +195,16 @@ def cc_CallableType : CCType<"Callable", "callable"> {
 }
 
 def cc_IndirectCallableType : CCType<"IndirectCallable", "indirect_callable"> {
-  let summary = "Proxy for cudaq::qkernel_ref.";
+  let summary = "Proxy for cudaq::qkernel.";
   let description = [{
     An entry-point kernel may take a reference to another kernel as an argument.
     The passed kernel may be entirely opaque at compile-time with its definition
     present in some other compilation module.
 
-    It is on the programmer to use the cudaq::qkernel_ref type. This wrapper
-    class is very much like std::function, but it extends that functionality
-    with some extra information for the runtime to be able to "link" the
-    distinct kernels on the device side and provide, for example, LTO at
-    JIT compile time.
+    It is on the programmer to use the cudaq::qkernel type. This wrapper class
+    is very much like std::function, but it extends that functionality with some
+    extra information for the runtime to be able to "link" the distinct kernels
+    on the device side and provide, for example, LTO at JIT compile time.
   }];
 
   let parameters = (ins "mlir::FunctionType":$signature);
diff --git a/lib/Frontend/nvqpp/ConvertDecl.cpp b/lib/Frontend/nvqpp/ConvertDecl.cpp
index 39ef0ad8df..7d277197a8 100644
--- a/lib/Frontend/nvqpp/ConvertDecl.cpp
+++ b/lib/Frontend/nvqpp/ConvertDecl.cpp
@@ -177,7 +177,7 @@ bool QuakeBridgeVisitor::interceptRecordDecl(clang::RecordDecl *x) {
       return pushType(cc::StateType::get(ctx));
     if (name.equals("pauli_word"))
       return pushType(cc::CharspanType::get(ctx));
-    if (name.equals("qkernel_ref")) {
+    if (name.equals("qkernel")) {
       auto *cts = cast<clang::ClassTemplateSpecializationDecl>(x);
       // Traverse template argument 0 to get the function's signature.
       if (!TraverseType(cts->getTemplateArgs()[0].getAsType()))
diff --git a/runtime/cudaq/cudaq.cpp b/runtime/cudaq/cudaq.cpp
index 4e4b29cd73..9373d05139 100644
--- a/runtime/cudaq/cudaq.cpp
+++ b/runtime/cudaq/cudaq.cpp
@@ -14,7 +14,7 @@
 #include "cuda_runtime_api.h"
 #endif
 #include "cudaq/platform.h"
-#include "cudaq/qis/qkernel_ref.h"
+#include "cudaq/qis/qkernel.h"
 #include "cudaq/utils/registry.h"
 #include "distributed/mpi_plugin.h"
 #include <dlfcn.h>
@@ -253,7 +253,7 @@ void cudaq::registry::__cudaq_registerLinkableKernel(void *hostSideFunc,
 std::intptr_t cudaq::registry::__cudaq_getLinkableKernelKey(void *p) {
   if (!p)
     throw std::runtime_error("cannot get kernel key, nullptr");
-  const auto &qk = *reinterpret_cast<const cudaq::qkernel_ref<void()> *>(p);
+  const auto &qk = *reinterpret_cast<const cudaq::qkernel<void()> *>(p);
   return reinterpret_cast<std::intptr_t>(*qk.get_entry_kernel_from_holder());
 }
 
diff --git a/runtime/cudaq/qis/qkernel_ref.h b/runtime/cudaq/qis/qkernel.h
similarity index 85%
rename from runtime/cudaq/qis/qkernel_ref.h
rename to runtime/cudaq/qis/qkernel.h
index f7e50223f5..6f477110fe 100644
--- a/runtime/cudaq/qis/qkernel_ref.h
+++ b/runtime/cudaq/qis/qkernel.h
@@ -17,9 +17,9 @@
 namespace cudaq {
 
 /// In library mode, the quake compiler is not involved. To streamline things,
-/// just have qkernel_ref alias the std::function template class.
+/// just have `qkernel` alias the std::function template class.
 template <typename Sig>
-using qkernel_ref = std::function<Sig>;
+using qkernel = std::function<Sig>;
 
 } // namespace cudaq
 
@@ -100,7 +100,7 @@ template <typename A>
 using remove_cvref_t = std::remove_cv_t<std::remove_reference_t<A>>;
 #endif
 
-/// A qkernel_ref<> must be used to wrap `CUDA-Q` kernels (callables annotated
+/// A `qkernel` must be used to wrap `CUDA-Q` kernels (callables annotated
 /// with the `__qpu__` attribute) when those kernels are \e referenced other
 /// than by a direct call in code outside of quantum kernels proper. Supports
 /// free functions, classes with call operators, and lambdas.
@@ -110,18 +110,18 @@ using remove_cvref_t = std::remove_cv_t<std::remove_reference_t<A>>;
 /// either stitch together execution in a simulation environment and/or JIT
 /// compile and re-link these kernels into a cohesive quantum circuit.
 template <typename>
-class qkernel_ref;
+class qkernel;
 
 template <typename R, typename... As>
-class qkernel_ref<R(As...)> {
+class qkernel<R(As...)> {
 public:
-  qkernel_ref() {}
-  qkernel_ref(std::nullptr_t) {}
-  qkernel_ref(const qkernel_ref &) = default;
-  qkernel_ref(qkernel_ref &&) = default;
+  qkernel() {}
+  qkernel(std::nullptr_t) {}
+  qkernel(const qkernel &) = default;
+  qkernel(qkernel &&) = default;
 
   template <typename FUNC,
-            bool SELF = std::is_same_v<remove_cvref_t<FUNC>, qkernel_ref>>
+            bool SELF = std::is_same_v<remove_cvref_t<FUNC>, qkernel>>
   using DecayType = typename std::enable_if_t<!SELF, std::decay<FUNC>>::type;
 
   template <typename FUNC, typename DFUNC = DecayType<FUNC>,
@@ -131,7 +131,7 @@ class qkernel_ref<R(As...)> {
   struct CallableType : RES {};
 
   template <typename S, typename = std::enable_if_t<CallableType<S>::value>>
-  qkernel_ref(S &&f) {
+  qkernel(S &&f) {
     using PS = remove_cvref_t<S>;
     if constexpr (std::is_same_v<PS, R (*)(As...)> ||
                   std::is_same_v<PS, R(As...)>) {
@@ -162,31 +162,31 @@ class qkernel_ref<R(As...)> {
 // Deduction guides for C++20.
 
 template <typename R, typename... As>
-qkernel_ref(R (*)(As...)) -> qkernel_ref<R(As...)>;
+qkernel(R (*)(As...)) -> qkernel<R(As...)>;
 
 template <typename>
-struct qkernel_ref_deduction_guide_helper {};
+struct qkernel_deduction_guide_helper {};
 
 template <typename R, typename P, typename... As>
-struct qkernel_ref_deduction_guide_helper<R (P::*)(As...)> {
+struct qkernel_deduction_guide_helper<R (P::*)(As...)> {
   using type = R(As...);
 };
 template <typename R, typename P, typename... As>
-struct qkernel_ref_deduction_guide_helper<R (P::*)(As...) const> {
+struct qkernel_deduction_guide_helper<R (P::*)(As...) const> {
   using type = R(As...);
 };
 template <typename R, typename P, typename... As>
-struct qkernel_ref_deduction_guide_helper<R (P::*)(As...) &> {
+struct qkernel_deduction_guide_helper<R (P::*)(As...) &> {
   using type = R(As...);
 };
 template <typename R, typename P, typename... As>
-struct qkernel_ref_deduction_guide_helper<R (P::*)(As...) const &> {
+struct qkernel_deduction_guide_helper<R (P::*)(As...) const &> {
   using type = R(As...);
 };
 
-template <typename F, typename S = typename qkernel_ref_deduction_guide_helper<
+template <typename F, typename S = typename qkernel_deduction_guide_helper<
                           decltype(&F::operator())>::type>
-qkernel_ref(F) -> qkernel_ref<S>;
+qkernel(F) -> qkernel<S>;
 
 #endif // CUDAQ_USE_STD20
 
diff --git a/runtime/cudaq/qis/qubit_qis.h b/runtime/cudaq/qis/qubit_qis.h
index 5e09625d1c..c83dffe844 100644
--- a/runtime/cudaq/qis/qubit_qis.h
+++ b/runtime/cudaq/qis/qubit_qis.h
@@ -13,7 +13,7 @@
 #include "cudaq/qis/modifiers.h"
 #include "cudaq/qis/pauli_word.h"
 #include "cudaq/qis/qarray.h"
-#include "cudaq/qis/qkernel_ref.h"
+#include "cudaq/qis/qkernel.h"
 #include "cudaq/qis/qreg.h"
 #include "cudaq/qis/qvector.h"
 #include "cudaq/spin_op.h"
diff --git a/runtime/cudaq/utils/registry.h b/runtime/cudaq/utils/registry.h
index a9b96bb3c0..3e339c196b 100644
--- a/runtime/cudaq/utils/registry.h
+++ b/runtime/cudaq/utils/registry.h
@@ -19,7 +19,7 @@ void cudaqRegisterLambdaName(const char *, const char *);
 /// Register a kernel with the runtime for kernel runtime stitching.
 void __cudaq_registerLinkableKernel(void *, const char *, void *);
 
-/// Return the kernel key from a qkernel_ref object. If \p p is a `nullptr` this
+/// Return the kernel key from a `qkernel` object. If \p p is a `nullptr` this
 /// will throw a runtime error.
 std::intptr_t __cudaq_getLinkableKernelKey(void *p);
 
diff --git a/targettests/SeparateCompilation/basic.cpp b/targettests/SeparateCompilation/basic.cpp
index 81f429a47c..eae9d13f3d 100644
--- a/targettests/SeparateCompilation/basic.cpp
+++ b/targettests/SeparateCompilation/basic.cpp
@@ -40,8 +40,7 @@ __qpu__ void dunkadee(cudaq::qvector<> &q) {
 #include "baselib.h"
 #include <iostream>
 
-__qpu__ void
-userKernel(const cudaq::qkernel_ref<void(cudaq::qvector<> &)> &init) {
+__qpu__ void userKernel(const cudaq::qkernel<void(cudaq::qvector<> &)> &init) {
   cudaq::qvector q(2);
   init(q);
 }
diff --git a/targettests/SeparateCompilation/class.cpp b/targettests/SeparateCompilation/class.cpp
index 87828bba20..eec15ccedd 100644
--- a/targettests/SeparateCompilation/class.cpp
+++ b/targettests/SeparateCompilation/class.cpp
@@ -42,8 +42,7 @@ void HereIsTheThing::operator()(cudaq::qvector<> &q) __qpu__ {
 #include "classlib.h"
 #include <iostream>
 
-__qpu__ void
-userKernel(const cudaq::qkernel_ref<void(cudaq::qvector<> &)> &init) {
+__qpu__ void userKernel(const cudaq::qkernel<void(cudaq::qvector<> &)> &init) {
   cudaq::qvector q(2);
   init(q);
 }
diff --git a/targettests/SeparateCompilation/deduction_guide.cpp b/targettests/SeparateCompilation/deduction_guide.cpp
index d585be56fa..9f64c5e5d6 100644
--- a/targettests/SeparateCompilation/deduction_guide.cpp
+++ b/targettests/SeparateCompilation/deduction_guide.cpp
@@ -34,8 +34,7 @@ __qpu__ void dunkadee(cudaq::qvector<> &q) { x(q[0]); }
 #include "udedgulib.h"
 #include <iostream>
 
-__qpu__ void
-userKernel(const cudaq::qkernel_ref<void(cudaq::qvector<> &)> &init) {
+__qpu__ void userKernel(const cudaq::qkernel<void(cudaq::qvector<> &)> &init) {
   cudaq::qvector q(2);
   init(q);
 }
diff --git a/targettests/SeparateCompilation/emulate.cpp b/targettests/SeparateCompilation/emulate.cpp
index 1b9def0bab..9128a93548 100644
--- a/targettests/SeparateCompilation/emulate.cpp
+++ b/targettests/SeparateCompilation/emulate.cpp
@@ -34,15 +34,14 @@ __qpu__ void dunkadee(cudaq::qvector<> &q) { x(q[0]); }
 #include "emulib.h"
 #include <iostream>
 
-__qpu__ void
-userKernel(const cudaq::qkernel_ref<void(cudaq::qvector<> &)> &init) {
+__qpu__ void userKernel(const cudaq::qkernel<void(cudaq::qvector<> &)> &init) {
   cudaq::qvector q(2);
   init(q);
 }
 
 int main() {
   cudaq::sample(10, userKernel,
-                cudaq::qkernel_ref<void(cudaq::qvector<> &)>{dunkadee});
+                cudaq::qkernel<void(cudaq::qvector<> &)>{dunkadee});
   std::cout << "Hello, World!\n";
   return 0;
 }
diff --git a/targettests/SeparateCompilation/lambda.cpp b/targettests/SeparateCompilation/lambda.cpp
index 5fa3b301b1..724d81d804 100644
--- a/targettests/SeparateCompilation/lambda.cpp
+++ b/targettests/SeparateCompilation/lambda.cpp
@@ -21,14 +21,13 @@
 
 #include "cudaq.h"
 
-__qpu__ void userKernel(const cudaq::qkernel_ref<void(cudaq::qvector<> &)> &);
+__qpu__ void userKernel(const cudaq::qkernel<void(cudaq::qvector<> &)> &);
 
 //--- anonlib.cpp
 
 #include "anonlib.h"
 
-__qpu__ void
-userKernel(const cudaq::qkernel_ref<void(cudaq::qvector<> &)> &init) {
+__qpu__ void userKernel(const cudaq::qkernel<void(cudaq::qvector<> &)> &init) {
   cudaq::qvector q(2);
   init(q);
 }
diff --git a/targettests/SeparateCompilation/multiple_callables.cpp b/targettests/SeparateCompilation/multiple_callables.cpp
index 8ed4a02b65..2f8bc5e6a1 100644
--- a/targettests/SeparateCompilation/multiple_callables.cpp
+++ b/targettests/SeparateCompilation/multiple_callables.cpp
@@ -11,8 +11,8 @@
 
 #include "cudaq.h"
 
-__qpu__ void entry(const cudaq::qkernel_ref<void(cudaq::qvector<> &)> &o,
-                   const cudaq::qkernel_ref<void(cudaq::qvector<> &, int)> &p) {
+__qpu__ void entry(const cudaq::qkernel<void(cudaq::qvector<> &)> &o,
+                   const cudaq::qkernel<void(cudaq::qvector<> &, int)> &p) {
   cudaq::qvector q(2);
   o(q);
   p(q, 1);
@@ -25,4 +25,3 @@ int main() {
   entry(l, m);
   return 0;
 }
-
diff --git a/targettests/execution/cudaq_observe.cpp b/targettests/execution/cudaq_observe.cpp
index c983ce36d2..daa10c1425 100644
--- a/targettests/execution/cudaq_observe.cpp
+++ b/targettests/execution/cudaq_observe.cpp
@@ -8,7 +8,6 @@
 
 // REQUIRES: c++20
 // clang-format off
-// RUN: nvq++ --target anyon                    --emulate %s -o %t && %t | FileCheck %s
 // RUN: nvq++ --target ionq                     --emulate %s -o %t && %t | FileCheck %s
 // 2 different IQM machines for 2 different topologies
 // RUN: nvq++ --target iqm --iqm-machine Adonis --emulate %s -o %t && %t | FileCheck %s
diff --git a/test/AST-Quake/indirect_callable.cpp b/test/AST-Quake/indirect_callable.cpp
index 5fb4af1106..ba429e9782 100644
--- a/test/AST-Quake/indirect_callable.cpp
+++ b/test/AST-Quake/indirect_callable.cpp
@@ -12,19 +12,19 @@
 
 __qpu__ int rando_qernel(double);
 
-__qpu__ void superstar_qernel(const cudaq::qkernel_ref<int(double)>& bob, double dub) {
+__qpu__ void superstar_qernel(const cudaq::qkernel<int(double)>& bob, double dub) {
    auto size = bob(dub);
    cudaq::qvector q(size);
    mz(q);
 }
 
 void meanwhile_on_safari() {
-   cudaq::qkernel_ref<int(double)> tiger{rando_qernel};
+   cudaq::qkernel<int(double)> tiger{rando_qernel};
    superstar_qernel(tiger, 11.0);
 }
 
 // clang-format off
-// CHECK-LABEL:   func.func @__nvqpp__mlirgen__function_superstar_qernel._Z16superstar_qernelRKN5cudaq11qkernel_refIFidEEEd(
+// CHECK-LABEL:   func.func @__nvqpp__mlirgen__function_superstar_qernel._Z16superstar_qernelRKN5cudaq7qkernelIFidEEEd(
 // CHECK-SAME:      %[[VAL_0:.*]]: !cc.indirect_callable<(f64) -> i32>,
 // CHECK-SAME:      %[[VAL_1:.*]]: f64) attributes {"cudaq-entrypoint", "cudaq-kernel", no_this} {
 // CHECK:           %[[VAL_2:.*]] = cc.alloca f64
@@ -40,7 +40,7 @@ void meanwhile_on_safari() {
 // CHECK:           return
 // CHECK:         }
 
-// CHECK-LABEL:   func.func @_Z16superstar_qernelRKN5cudaq11qkernel_refIFidEEEd(
+// CHECK-LABEL:   func.func @_Z16superstar_qernelRKN5cudaq7qkernelIFidEEEd(
 // CHECK-SAME:      %[[VAL_0:.*]]: !cc.ptr<i8>,
 // CHECK-SAME:      %[[VAL_1:.*]]: f64) attributes {no_this} {
 // CHECK:           return