diff --git a/cpp/sopt/primal_dual.h b/cpp/sopt/primal_dual.h
index 2767bcca0..c643e1a7b 100644
--- a/cpp/sopt/primal_dual.h
+++ b/cpp/sopt/primal_dual.h
@@ -244,7 +244,7 @@ class PrimalDual {
   //!
   //! This function simplifies creating overloads for operator() in PD wrappers.
   static std::tuple<t_Vector, t_Vector> initial_guess(t_Vector const &target,
-                                                      t_LinearTransform const &phi, Real nu) {
+                                                      t_LinearTransform const &phi, Real  /*nu*/) {
     std::tuple<t_Vector, t_Vector> guess;
     std::get<0>(guess) = static_cast<t_Vector>(phi.adjoint() * target);
     std::get<1>(guess) = target;
diff --git a/cpp/sopt/proximal.h b/cpp/sopt/proximal.h
index a3d4e21e3..42bb364f8 100644
--- a/cpp/sopt/proximal.h
+++ b/cpp/sopt/proximal.h
@@ -132,7 +132,7 @@ void tv_norm(Eigen::DenseBase<T0> &out, Eigen::DenseBase<T2> const &gamma,
 
 //! Proximal of a function that is always zero, the identity
 template <class T0, class T1>
-void id(Eigen::DenseBase<T0> &out, typename real_type<typename T0::Scalar>::type gamma,
+void id(Eigen::DenseBase<T0> &out, typename real_type<typename T0::Scalar>::type  /*gamma*/,
         Eigen::DenseBase<T1> const &x) {
   out = x;
 }
diff --git a/cpp/sopt/tf_g_proximal.h b/cpp/sopt/tf_g_proximal.h
index ea6e7fb30..7aac7b885 100644
--- a/cpp/sopt/tf_g_proximal.h
+++ b/cpp/sopt/tf_g_proximal.h
@@ -63,7 +63,7 @@ class TFGProximal : public GProximal<SCALAR> {
 
   // Return g_proximal as a lambda function. Used in operator() in base class.
   t_Proximal proximal_function() const override {
-    return [this](t_Vector &out, Real gamma, t_Vector const &x) {
+    return [this](t_Vector &out, Real  /*gamma*/, t_Vector const &x) {
 	     this -> call_model(out, x);
 	   };
   }
diff --git a/cpp/sopt/wavelets.h b/cpp/sopt/wavelets.h
index 197d1c427..8ceb78d45 100644
--- a/cpp/sopt/wavelets.h
+++ b/cpp/sopt/wavelets.h
@@ -108,7 +108,7 @@ Vector<T> &get_wavelet_basis_coefficients(Vector<T> &coeffs, const t_uint basis_
   return coeffs.segment(basis_index * size, size);
 }
 template <class T>
-Vector<T> &get_wavelet_levels_1d(Vector<T> &coeffs, const t_uint level, const t_uint size) {
+Vector<T> &get_wavelet_levels_1d(Vector<T> &coeffs, const t_uint level, const t_uint  /*size*/) {
   auto const N = static_cast<t_uint>(coeffs.size()) >> level;  // bitshift to divide by 2^level
   return coeffs.head(N);
 }
diff --git a/cpp/tests/credible_region.cc b/cpp/tests/credible_region.cc
index d50b0d225..70ae272d1 100644
--- a/cpp/tests/credible_region.cc
+++ b/cpp/tests/credible_region.cc
@@ -14,7 +14,7 @@ t_uint N = rows * cols;
 
 TEST_CASE("calculating gamma") {
   sopt::logging::set_level("debug");
-  const std::function<t_real(t_Vector)> energy_function = [](const t_Vector &input) -> t_real {
+  const std::function<t_real(t_Vector)> energy_function = [](const t_Vector & /*input*/) -> t_real {
     return 0.;
   };
   const t_Vector x = t_Vector::Random(N);
diff --git a/cpp/tests/forward_backward.cc b/cpp/tests/forward_backward.cc
index ab1c195bd..8b2d2f43a 100644
--- a/cpp/tests/forward_backward.cc
+++ b/cpp/tests/forward_backward.cc
@@ -39,7 +39,7 @@ TEST_CASE("Forward Backward with ||x - x0||_2^2 function", "[fb]") {
   auto const grad = [](t_Vector &out, const t_Vector &x) { out = x; };
   const t_Vector x_guess = t_Vector::Random(target0.size());
   const t_Vector res = x_guess - target0;
-  auto const convergence = [&target0](const t_Vector &x, const t_Vector &res) -> bool {
+  auto const convergence = [&target0](const t_Vector &x, const t_Vector & /*res*/) -> bool {
     return x.isApprox(target0, 1e-9);
   };
   CAPTURE(target0);
diff --git a/cpp/tests/primal_dual.cc b/cpp/tests/primal_dual.cc
index 68cdb957c..41c438239 100644
--- a/cpp/tests/primal_dual.cc
+++ b/cpp/tests/primal_dual.cc
@@ -60,7 +60,7 @@ TEST_CASE("Primal Dual with 0.5 * ||x - x0||_2^2 function", "[primaldual]") {
   auto const g = proximal::L2Norm<Scalar>();
   const t_Vector x_guess = t_Vector::Random(target0.size());
   const t_Vector res = x_guess - target0;
-  auto const convergence = [&target0](const t_Vector &x, const t_Vector &res) -> bool {
+  auto const convergence = [&target0](const t_Vector &x, const t_Vector & /*res*/) -> bool {
     return x.isApprox(target0, 1e-9);
   };
   CAPTURE(target0);