diff --git a/.github/workflows/benchmark.yml b/.github/workflows/benchmark.yml
new file mode 100644
index 0000000..e0d470e
--- /dev/null
+++ b/.github/workflows/benchmark.yml
@@ -0,0 +1,16 @@
+# https://astroautomata.com/AirspeedVelocity.jl/stable/
+
+name: Benchmark
+
+on: [pull_request, workflow_dispatch]
+
+permissions:
+  pull-requests: write  # need to post comments
+
+jobs:
+  bench:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: MilesCranmer/AirspeedVelocity.jl@action-v1
+        with:
+          julia-version: '1'
diff --git a/Project.toml b/Project.toml
index 9faf74f..5f22f3e 100644
--- a/Project.toml
+++ b/Project.toml
@@ -6,7 +6,6 @@ version = "1.0.0-DEV"
 [deps]
 AbstractTrees = "1520ce14-60c1-5f80-bbc7-55ef81b5835c"
 AllocCheck = "9b6a8646-10ed-4001-bbdc-1d2f46dfbb1a"
-BenchmarkTools = "6e4b80f9-dd63-53aa-95a3-0cdb28fa8baf"
 BlockArrays = "8e7c35d0-a365-5155-bbbb-fb81a777f24e"
 DrWatson = "634d3b9d-ee7a-5ddf-bec9-22491ea816e1"
 ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210"
@@ -24,6 +23,5 @@ Profile = "9abbd945-dff8-562f-b5e8-e1ebf5ef1b79"
 ProfileView = "c46f51b8-102a-5cf2-8d2c-8597cb0e0da7"
 Roots = "f2b01f46-fcfa-551c-844a-d8ac1e96c665"
 StaticArrays = "90137ffa-7385-5640-81b9-e52037218182"
-Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
 TimerOutputs = "a759f4b9-e2f1-59dc-863e-4aeb61b1ea8f"
 WriteVTK = "64499a7a-5c06-52f2-abe2-ccb03c286192"
diff --git a/test/Project.toml b/test/Project.toml
new file mode 100644
index 0000000..42e3f98
--- /dev/null
+++ b/test/Project.toml
@@ -0,0 +1,6 @@
+[deps]
+ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210"
+Gridap = "56d4f2e9-7ea1-5844-9cf6-b9c51ca7ce8e"
+JSON = "682c06a0-de6a-54ab-a142-c8b1cf79cde6"
+StaticArrays = "90137ffa-7385-5640-81b9-e52037218182"
+Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
diff --git a/test/TestConstitutiveModels/PhysicalModels.jl b/test/TestConstitutiveModels/PhysicalModels.jl
index 6e411d7..5ad93f0 100644
--- a/test/TestConstitutiveModels/PhysicalModels.jl
+++ b/test/TestConstitutiveModels/PhysicalModels.jl
@@ -4,7 +4,6 @@ using JSON
 using StaticArrays
 using Test
 using HyperFEM.PhysicalModels
-using BenchmarkTools
 
 
 import Base: -
@@ -39,31 +38,6 @@ function test_derivatives_3D_(model::PhysicalModel; rtol=1e-14, kwargs...)
 end
 
 
-function benchmark_derivatives__(model::PhysicalModel, ∇u)
-  Ψ, ∂Ψu, ∂Ψuu = model()
-  ∂Ψu_(F) = TensorValue(ForwardDiff.gradient(Ψ, get_array(F)))
-  ∂Ψuu_(F) = TensorValue(ForwardDiff.hessian(Ψ, get_array(F)))
-
-  F, _, _ = get_Kinematics(model.Kinematic)
-  print("Analyitical ∂Ψu  | ")
-  @btime $∂Ψu($F($∇u))
-  print("Numerical ∂Ψu    | ")
-  @btime $∂Ψu_($F($∇u))
-  print("Analyitical ∂Ψuu | ")
-  @btime $∂Ψuu($F($∇u))
-  print("Numerical ∂Ψuu   | ")
-  @btime $∂Ψuu_($F($∇u))
-end
-
-function benchmark_derivatives_2D_(model::PhysicalModel)
-  benchmark_derivatives__(model, ∇u2)
-end
-
-function benchmark_derivatives_3D_(model::PhysicalModel)
-  benchmark_derivatives__(model, ∇u3)
-end
-
-
 @testset "NonlinearMooneyRivlin_CV" begin
   model = NonlinearMooneyRivlin_CV(λ=3.0, μ1=1.0, μ2=1.0, α=2.0, β=1.0, γ=6.0)
   test_derivatives_3D_(model, rtol=1e-13)
diff --git a/test/TestConstitutiveModels/runtests.jl b/test/TestConstitutiveModels/runtests.jl
index ff18c32..87d5273 100644
--- a/test/TestConstitutiveModels/runtests.jl
+++ b/test/TestConstitutiveModels/runtests.jl
@@ -1,6 +1,6 @@
 using HyperFEM
 
-@testset "ConstitutiveModels" verbose = true begin
+@testset "ConstitutiveModels" begin
 
     @time begin
         include("PhysicalModels.jl")
diff --git a/test/TestTensorAlgebra/TensorAlgebra.jl b/test/TestTensorAlgebra/TensorAlgebra.jl
index 22f0fcf..126ecfc 100644
--- a/test/TestTensorAlgebra/TensorAlgebra.jl
+++ b/test/TestTensorAlgebra/TensorAlgebra.jl
@@ -2,7 +2,6 @@ using Gridap.TensorValues
 using Gridap.Arrays
 using HyperFEM.TensorAlgebra
 using Test
-using BenchmarkTools
 
 
 @testset "Jacobian regularization" begin
@@ -31,21 +30,6 @@ end
 end
 
 
-# @benchmark (A ⊗₁₃²⁴ B)
-# @benchmark (A ⊗₁₂³ V1)
-# @benchmark (A ⊗₁₃² V1)
-# @benchmark (V1 ⊗₁²³ A)
-# @benchmark (A ⊗₁₃²⁴ B)
-# @benchmark (D × A)
-
-# @code_warntype (A ⊗₁₃²⁴ B)
-# @code_warntype (A ⊗₁₂³ V1)
-# @code_warntype (A ⊗₁₃² V1)
-# @code_warntype (V1 ⊗₁²³ A)
-# @code_warntype (A ⊗₁₃²⁴ B)
-# @code_warntype (D × A)
-
-
 @testset "cross" begin
   A = TensorValue(1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0) * 1e-3
   B = TensorValue(4.6, 2.1, 1.7, 3.2, 6.5, 1.4, 9.2, 8.0, 9.0) * 1e-3
@@ -106,10 +90,6 @@ end
   @test I3_ == I3
   @test I4_ == I4
   @test I9_ == I9
-  # @benchmark I2_
-  # @benchmark I2
-  # @benchmark I9_
-  # @benchmark I9
 end
 
 
diff --git a/test/TestTensorAlgebra/runtests.jl b/test/TestTensorAlgebra/runtests.jl
index f1e4b72..7b354b0 100644
--- a/test/TestTensorAlgebra/runtests.jl
+++ b/test/TestTensorAlgebra/runtests.jl
@@ -1,4 +1,4 @@
-@testset "TensorAlgebra" verbose = true begin
+@testset "TensorAlgebra" begin
 
     @time begin
         include("TensorAlgebra.jl")
diff --git a/test/TestWeakForms/runtests.jl b/test/TestWeakForms/runtests.jl
index 72a7843..29cd81b 100644
--- a/test/TestWeakForms/runtests.jl
+++ b/test/TestWeakForms/runtests.jl
@@ -1,4 +1,4 @@
-@testset "WeakForms" verbose = true begin
+@testset "WeakForms" begin
 
     @time begin
         include("WeakForms.jl")
diff --git a/test/runtests.jl b/test/runtests.jl
index d7188ba..9d9fc57 100644
--- a/test/runtests.jl
+++ b/test/runtests.jl
@@ -1,13 +1,6 @@
 using HyperFEM
 using Test
-using Profile
-using Gridap, GridapGmsh, GridapSolvers, DrWatson, TimerOutputs
-using GridapSolvers.NonlinearSolvers
-using Gridap.FESpaces
-using Gridap.CellData
-using Gridap.TensorValues
-using HyperFEM: jacobian, IterativeSolver, solve!
-using BenchmarkTools
+using Gridap
 
 
 @testset "HyperFEMTests" verbose = true begin
@@ -23,4 +16,3 @@ using BenchmarkTools
     end
 
 end;
-