diff --git a/Cargo.toml b/Cargo.toml
index 91fd9c05..2729fafc 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -58,4 +58,8 @@ harness = false
 [[bench]]
 name = "bn256_field"
 harness = false
-required-features = ["reexport"]
\ No newline at end of file
+required-features = ["reexport"]
+
+[[bench]]
+name = "group"
+harness = false
diff --git a/benches/group.rs b/benches/group.rs
new file mode 100644
index 00000000..68cfee53
--- /dev/null
+++ b/benches/group.rs
@@ -0,0 +1,52 @@
+use criterion::{black_box, criterion_group, criterion_main, Criterion};
+use ff::Field;
+use group::prime::PrimeCurveAffine;
+use halo2curves::secp256k1::Secp256k1;
+use pasta_curves::arithmetic::CurveExt;
+use rand_core::OsRng;
+
+fn criterion_benchmark<G: CurveExt>(c: &mut Criterion) {
+    // G1Projective
+    {
+        let name = "GProjective";
+        let p1 = G::random(OsRng);
+        let p2 = G::random(OsRng);
+        let p1_affine = G::AffineExt::from(p1);
+        let s = G::ScalarExt::random(OsRng);
+
+        const N: usize = 1000;
+        let v = vec![G::generator(); N];
+        let mut q = vec![G::AffineExt::identity(); N];
+
+        c.bench_function(&format!("{} check on curve", name), move |b| {
+            b.iter(|| black_box(p1).is_on_curve())
+        });
+        c.bench_function(&format!("{} check equality", name), move |b| {
+            b.iter(|| black_box(p1) == black_box(p1))
+        });
+        c.bench_function(&format!("{} to affine", name), move |b| {
+            b.iter(|| G::AffineExt::from(black_box(p1)))
+        });
+        c.bench_function(&format!("{} doubling", name), move |b| {
+            b.iter(|| black_box(p1).double())
+        });
+        c.bench_function(&format!("{} addition", name), move |b| {
+            b.iter(|| black_box(p1).add(&p2))
+        });
+        c.bench_function(&format!("{} mixed addition", name), move |b| {
+            b.iter(|| black_box(p2).add(&p1_affine))
+        });
+        c.bench_function(&format!("{} scalar multiplication", name), move |b| {
+            b.iter(|| black_box(p1) * black_box(s))
+        });
+        c.bench_function(&format!("{} batch to affine n={}", name, N), move |b| {
+            b.iter(|| {
+                G::batch_normalize(black_box(&v), black_box(&mut q));
+                black_box(&q)[0]
+            })
+        });
+    }
+}
+
+criterion_group!(benches, criterion_benchmark<Secp256k1>);
+criterion_main!(benches);