diff --git a/ext/MathOptAIFluxExt.jl b/ext/MathOptAIFluxExt.jl
index c3df772..94e940b 100644
--- a/ext/MathOptAIFluxExt.jl
+++ b/ext/MathOptAIFluxExt.jl
@@ -141,6 +141,10 @@ function MathOptAI.build_predictor(
     return inner_predictor
 end
 
+function _add_predictor(::MathOptAI.Pipeline, layer::Any, ::Dict)
+    return error("Unsupported layer: $layer")
+end
+
 _default(::typeof(identity)) = nothing
 _default(::Any) = missing
 _default(::typeof(Flux.relu)) = MathOptAI.ReLU()
diff --git a/ext/MathOptAILuxExt.jl b/ext/MathOptAILuxExt.jl
index 57dc99f..9690337 100644
--- a/ext/MathOptAILuxExt.jl
+++ b/ext/MathOptAILuxExt.jl
@@ -160,6 +160,10 @@ function MathOptAI.build_predictor(
     return inner_predictor
 end
 
+function _add_predictor(::MathOptAI.Pipeline, layer::Any, ::Any, ::Dict)
+    return error("Unsupported layer: $layer")
+end
+
 _default(::typeof(identity)) = nothing
 _default(::Any) = missing
 _default(::typeof(Lux.relu)) = MathOptAI.ReLU()
@@ -182,7 +186,7 @@ end
 function _add_predictor(
     predictor::MathOptAI.Pipeline,
     layer::Lux.Dense,
-    p,
+    p::Any,
     config::Dict,
 )
     push!(predictor.layers, MathOptAI.Affine(p.weight, vec(p.bias)))
@@ -193,7 +197,7 @@ end
 function _add_predictor(
     predictor::MathOptAI.Pipeline,
     layer::Lux.Scale,
-    p,
+    p::Any,
     config::Dict,
 )
     push!(predictor.layers, MathOptAI.Scale(p.weight, p.bias))
diff --git a/test/test_Flux.jl b/test/test_Flux.jl
index 2b9c3d4..86088a3 100644
--- a/test/test_Flux.jl
+++ b/test/test_Flux.jl
@@ -190,6 +190,17 @@ function test_end_to_end_Tanh()
     return
 end
 
+function test_unsupported_layer()
+    layer = Flux.Conv((5, 5), 3 => 7)
+    model = Model()
+    @variable(model, x[1:2])
+    @test_throws(
+        ErrorException("Unsupported layer: $layer"),
+        MathOptAI.add_predictor(model, Flux.Chain(layer), x),
+    )
+    return
+end
+
 end  # module
 
 TestFluxExt.runtests()
diff --git a/test/test_Lux.jl b/test/test_Lux.jl
index 42b79cf..38e507f 100644
--- a/test/test_Lux.jl
+++ b/test/test_Lux.jl
@@ -198,6 +198,20 @@ function test_end_to_end_Tanh()
     return
 end
 
+function test_unsupported_layer()
+    layer = Lux.Conv((5, 5), 3 => 7)
+    rng = Random.MersenneTwister()
+    ml_model = Lux.Chain(layer, layer)
+    parameters, state = Lux.setup(rng, ml_model)
+    model = Model()
+    @variable(model, x[1:2])
+    @test_throws(
+        ErrorException("Unsupported layer: $layer"),
+        MathOptAI.add_predictor(model, (ml_model, parameters, state), x),
+    )
+    return
+end
+
 end  # module
 
 TestLuxExt.runtests()