diff --git a/Project.toml b/Project.toml
index 3af67f51c..3e6a8b47d 100644
--- a/Project.toml
+++ b/Project.toml
@@ -1,7 +1,7 @@
 name = "Lux"
 uuid = "b2108857-7c20-44ae-9111-449ecde12c47"
 authors = ["Avik Pal <avikpal@mit.edu> and contributors"]
-version = "1.4.0"
+version = "1.4.1-DEV"
 
 [deps]
 ADTypes = "47edcb42-4c32-4615-8424-f2b9edc5f35b"
@@ -79,7 +79,7 @@ DispatchDoctor = "0.4.12"
 Enzyme = "0.13.16"
 EnzymeCore = "0.8.6"
 FastClosures = "0.3.2"
-Flux = "0.14.25"
+Flux = "0.15"
 ForwardDiff = "0.10.36"
 FunctionWrappers = "1.1.3"
 Functors = "0.5"
@@ -95,7 +95,7 @@ MacroTools = "0.5.13"
 Markdown = "1.10"
 NCCL = "0.1.1"
 NNlib = "0.9.24"
-Optimisers = "0.3.4, 0.4"
+Optimisers = "0.4.1"
 Preferences = "1.4.3"
 Random = "1.10"
 Reactant = "0.2.6"
@@ -107,7 +107,7 @@ SimpleChains = "0.4.7"
 Static = "1.1.1"
 StaticArraysCore = "1.4.3"
 Statistics = "1.10"
-Tracker = "0.2.36"
+Tracker = "0.2.37"
 WeightInitializers = "1"
 Zygote = "0.6.70"
 julia = "1.10"
diff --git a/docs/Project.toml b/docs/Project.toml
index ff948b0b3..2294d0bfd 100644
--- a/docs/Project.toml
+++ b/docs/Project.toml
@@ -52,7 +52,7 @@ LuxLib = "1.3.4"
 LuxTestUtils = "1.5"
 MLDataDevices = "1.6"
 NNlib = "0.9.24"
-Optimisers = "0.3.4, 0.4"
+Optimisers = "0.4.1"
 Pkg = "1.10"
 Printf = "1.10"
 Random = "1.10"
diff --git a/examples/Basics/Project.toml b/examples/Basics/Project.toml
index 1e32e0ea0..c7ea884bf 100644
--- a/examples/Basics/Project.toml
+++ b/examples/Basics/Project.toml
@@ -13,5 +13,5 @@ ComponentArrays = "0.15.18"
 ForwardDiff = "0.10"
 Lux = "1"
 LuxCUDA = "0.3"
-Optimisers = "0.3.4, 0.4"
+Optimisers = "0.4.1"
 Zygote = "0.6"
diff --git a/examples/BayesianNN/Project.toml b/examples/BayesianNN/Project.toml
index f1a9d476f..7beb286cc 100644
--- a/examples/BayesianNN/Project.toml
+++ b/examples/BayesianNN/Project.toml
@@ -14,6 +14,6 @@ Functors = "0.4, 0.5"
 LinearAlgebra = "1"
 Lux = "1.2"
 Random = "1"
-Tracker = "0.2.36"
+Tracker = "0.2.37"
 Turing = "0.34, 0.35"
 Zygote = "0.6.69"
diff --git a/examples/ConvMixer/Project.toml b/examples/ConvMixer/Project.toml
index d1ffac2cd..8ae780657 100644
--- a/examples/ConvMixer/Project.toml
+++ b/examples/ConvMixer/Project.toml
@@ -31,7 +31,7 @@ LuxCUDA = "0.3.2"
 MLDatasets = "0.7.14"
 MLUtils = "0.4.4"
 OneHotArrays = "0.2.5"
-Optimisers = "0.4"
+Optimisers = "0.4.1"
 PreferenceTools = "0.1.2"
 Printf = "1.10"
 ProgressBars = "1.5.1"
diff --git a/examples/DDIM/Project.toml b/examples/DDIM/Project.toml
index 4608d02b2..743d7fc1d 100644
--- a/examples/DDIM/Project.toml
+++ b/examples/DDIM/Project.toml
@@ -36,7 +36,7 @@ JLD2 = "0.4.48, 0.5"
 Lux = "1"
 LuxCUDA = "0.3"
 MLUtils = "0.4"
-Optimisers = "0.3, 0.4"
+Optimisers = "0.4.1"
 ParameterSchedulers = "0.4.1"
 ProgressBars = "1"
 Random = "1.10"
diff --git a/examples/HyperNet/Project.toml b/examples/HyperNet/Project.toml
index 501036806..965427971 100644
--- a/examples/HyperNet/Project.toml
+++ b/examples/HyperNet/Project.toml
@@ -21,7 +21,7 @@ LuxCUDA = "0.3"
 MLDatasets = "0.7"
 MLUtils = "0.4"
 OneHotArrays = "0.2.5"
-Optimisers = "0.3.4, 0.4"
+Optimisers = "0.4.1"
 Setfield = "1"
 Statistics = "1"
 Zygote = "0.6"
diff --git a/examples/ImageNet/Project.toml b/examples/ImageNet/Project.toml
index 5a0c02255..7f91b02db 100644
--- a/examples/ImageNet/Project.toml
+++ b/examples/ImageNet/Project.toml
@@ -38,7 +38,7 @@ MLUtils = "0.4.4"
 MPI = "0.20.21"
 NCCL = "0.1.1"
 OneHotArrays = "0.2.5"
-Optimisers = "0.3.4, 0.4"
+Optimisers = "0.4.1"
 ParameterSchedulers = "0.4.2"
 Random = "1.10"
 Setfield = "1.1.1"
diff --git a/examples/NeuralODE/Project.toml b/examples/NeuralODE/Project.toml
index 9bb2456cc..a292da633 100644
--- a/examples/NeuralODE/Project.toml
+++ b/examples/NeuralODE/Project.toml
@@ -21,7 +21,7 @@ LuxCUDA = "0.3"
 MLDatasets = "0.7"
 MLUtils = "0.4"
 OneHotArrays = "0.2.5"
-Optimisers = "0.3.4, 0.4"
+Optimisers = "0.4.1"
 OrdinaryDiffEqTsit5 = "1"
 SciMLSensitivity = "7.63"
 Statistics = "1"
diff --git a/examples/PINN2DPDE/Project.toml b/examples/PINN2DPDE/Project.toml
index 57abc282e..4b2b24bf0 100644
--- a/examples/PINN2DPDE/Project.toml
+++ b/examples/PINN2DPDE/Project.toml
@@ -18,7 +18,7 @@ Lux = "1"
 LuxCUDA = "0.3.3"
 MLUtils = "0.4.4"
 OnlineStats = "1.7.1"
-Optimisers = "0.3.4, 0.4"
+Optimisers = "0.4.1"
 Printf = "1.10"
 Random = "1.10"
 Statistics = "1.10"
diff --git a/examples/PolynomialFitting/Project.toml b/examples/PolynomialFitting/Project.toml
index 36c4d4e59..168865fdd 100644
--- a/examples/PolynomialFitting/Project.toml
+++ b/examples/PolynomialFitting/Project.toml
@@ -14,6 +14,6 @@ ADTypes = "1.10"
 CairoMakie = "0.12"
 Lux = "1"
 LuxCUDA = "0.3"
-Optimisers = "0.3.4, 0.4"
+Optimisers = "0.4.1"
 Statistics = "1"
 Zygote = "0.6"
diff --git a/examples/SimpleChains/Project.toml b/examples/SimpleChains/Project.toml
index c8ec5a59b..3f1b9b2a4 100644
--- a/examples/SimpleChains/Project.toml
+++ b/examples/SimpleChains/Project.toml
@@ -16,7 +16,7 @@ Lux = "1"
 MLDatasets = "0.7.14"
 MLUtils = "0.4"
 OneHotArrays = "0.2.5"
-Optimisers = "0.3.4, 0.4"
+Optimisers = "0.4.1"
 Random = "1"
 SimpleChains = "0.4.6"
 Zygote = "0.6.69"
diff --git a/examples/SimpleRNN/Project.toml b/examples/SimpleRNN/Project.toml
index 78cbad4b6..81e54f61e 100644
--- a/examples/SimpleRNN/Project.toml
+++ b/examples/SimpleRNN/Project.toml
@@ -16,6 +16,6 @@ JLD2 = "0.5"
 Lux = "1"
 LuxCUDA = "0.3"
 MLUtils = "0.4"
-Optimisers = "0.3.4, 0.4"
+Optimisers = "0.4.1"
 Statistics = "1"
 Zygote = "0.6"
diff --git a/ext/LuxFluxExt.jl b/ext/LuxFluxExt.jl
index d0f89b2b0..f901823ed 100644
--- a/ext/LuxFluxExt.jl
+++ b/ext/LuxFluxExt.jl
@@ -211,12 +211,6 @@ function Lux.convert_flux_model(
     return Lux.GroupNorm(l.chs, l.G, l.λ; l.affine, epsilon=l.ϵ)
 end
 
-const _INVALID_TRANSFORMATION_TYPES = Union{<:Flux.Recur}
-
-function Lux.convert_flux_model(l::T; kwargs...) where {T <: _INVALID_TRANSFORMATION_TYPES}
-    throw(FluxModelConversionException("Transformation of type $(T) is not supported."))
-end
-
 for cell in (:RNNCell, :LSTMCell, :GRUCell)
     msg = "Recurrent Cell: $(cell) for Flux has semantical difference with Lux, \
            mostly in-terms of how the bias term is dealt with. Lux aligns with the Pytorch \
diff --git a/test/Project.toml b/test/Project.toml
index 429ba85c5..58dd94c2e 100644
--- a/test/Project.toml
+++ b/test/Project.toml
@@ -65,7 +65,7 @@ MLUtils = "0.4.3"
 NNlib = "0.9.24"
 Octavian = "0.3.28"
 OneHotArrays = "0.2.5"
-Optimisers = "0.3.4, 0.4"
+Optimisers = "0.4.1"
 Pkg = "1.10"
 Preferences = "1.4.3"
 Random = "1.10"
@@ -79,5 +79,5 @@ Static = "1"
 StaticArrays = "1.9"
 Statistics = "1.10"
 Test = "1.10"
-Tracker = "0.2.36"
+Tracker = "0.2.37"
 Zygote = "0.6.70"
diff --git a/test/runtests.jl b/test/runtests.jl
index 130ea0275..6837b9ae0 100644
--- a/test/runtests.jl
+++ b/test/runtests.jl
@@ -26,8 +26,7 @@ if ("all" in LUX_TEST_GROUP || "misc" in LUX_TEST_GROUP)
     push!(EXTRA_PKGS, Pkg.PackageSpec("MPI"))
     (BACKEND_GROUP == "all" || BACKEND_GROUP == "cuda") &&
         push!(EXTRA_PKGS, Pkg.PackageSpec("NCCL"))
-    # XXX: Reactivate once Flux is compatible with Functors 0.5
-    # push!(EXTRA_PKGS, Pkg.PackageSpec("Flux"))
+    push!(EXTRA_PKGS, Pkg.PackageSpec("Flux"))
 end
 
 if !Sys.iswindows()
diff --git a/test/transform/flux_tests.jl b/test/transform/flux_tests.jl
index 38bc1c176..6b9ab9d60 100644
--- a/test/transform/flux_tests.jl
+++ b/test/transform/flux_tests.jl
@@ -1,4 +1,4 @@
-@testitem "FromFluxAdaptor" setup=[SharedTestSetup] tags=[:misc] skip=:(true) begin
+@testitem "FromFluxAdaptor" setup=[SharedTestSetup] tags=[:misc] begin
     import Flux
 
     toluxpsst = FromFluxAdaptor(; preserve_ps_st=true)
@@ -8,10 +8,10 @@
     @testset "$mode" for (mode, aType, dev, ongpu) in MODES
         @testset "Containers" begin
             @testset "Chain" begin
-                models = [Flux.Chain(Flux.Dense(2 => 5), Flux.Dense(5 => 1)),
-                    Flux.Chain(; l1=Flux.Dense(2 => 5), l2=Flux.Dense(5 => 1))] |> dev
-
-                for model in models
+                for model in [
+                    Flux.Chain(Flux.Dense(2 => 5), Flux.Dense(5 => 1)) |> dev,
+                    Flux.Chain(; l1=Flux.Dense(2 => 5), l2=Flux.Dense(5 => 1)) |> dev
+                ]
                     x = rand(Float32, 2, 1) |> aType
 
                     model_lux = toluxpsst(model)
@@ -57,10 +57,10 @@
             end
 
             @testset "Parallel" begin
-                models = [Flux.Parallel(+, Flux.Dense(2 => 2), Flux.Dense(2 => 2)),
-                    Flux.Parallel(+; l1=Flux.Dense(2 => 2), l2=Flux.Dense(2 => 2))] |> dev
-
-                for model in models
+                for model in [
+                    Flux.Parallel(+, Flux.Dense(2 => 2), Flux.Dense(2 => 2)) |> dev,
+                    Flux.Parallel(+; l1=Flux.Dense(2 => 2), l2=Flux.Dense(2 => 2)) |> dev
+                ]
                     x = rand(Float32, 2, 1) |> aType
 
                     model_lux = toluxpsst(model)
@@ -94,8 +94,10 @@
 
         @testset "Linear" begin
             @testset "Dense" begin
-                for model in [Flux.Dense(2 => 4) |> dev,
-                    Flux.Dense(2 => 4; bias=false) |> dev]
+                for model in [
+                    Flux.Dense(2 => 4) |> dev,
+                    Flux.Dense(2 => 4; bias=false) |> dev
+                ]
                     x = randn(Float32, 2, 4) |> aType
 
                     model_lux = toluxpsst(model)
@@ -112,7 +114,9 @@
 
             @testset "Scale" begin
                 for model in [
-                    Flux.Scale(2) |> dev, Flux.Scale(2; bias=false) |> dev]
+                    Flux.Scale(2) |> dev,
+                    Flux.Scale(2; bias=false) |> dev
+                ]
                     x = randn(Float32, 2, 4) |> aType
 
                     model_lux = toluxpsst(model)
@@ -128,8 +132,10 @@
             end
 
             @testset "Bilinear" begin
-                for model in [Flux.Bilinear((2, 3) => 5) |> dev,
-                    Flux.Bilinear((2, 3) => 5; bias=false) |> dev]
+                for model in [
+                    Flux.Bilinear((2, 3) => 5) |> dev,
+                    Flux.Bilinear((2, 3) => 5; bias=false) |> dev
+                ]
                     x = randn(Float32, 2, 4) |> aType
                     y = randn(Float32, 3, 4) |> aType
 
@@ -447,7 +453,7 @@
                 bias
             end
 
-            Flux.@functor CustomFluxLayer
+            Flux.@layer CustomFluxLayer
 
             (c::CustomFluxLayer)(x) = c.weight .* x .+ c.bias
 
@@ -466,12 +472,5 @@
             @test tolux(identity) isa Lux.NoOpLayer
             @test tolux(+) isa Lux.WrappedFunction
         end
-
-        @testset "Unsupported Layers" begin
-            accum(h, x) = (h + x, x)
-            rnn = Flux.Recur(accum, 0)
-
-            @test_throws Lux.FluxModelConversionException tolux(rnn)
-        end
     end
 end