diff --git a/.github/workflows/benchmark.yml b/.github/workflows/benchmark.yml
new file mode 100644
index 0000000..4d37173
--- /dev/null
+++ b/.github/workflows/benchmark.yml
@@ -0,0 +1,127 @@
+# TODO(template) this workflow requires a `gh-pages` branch to exist.
+# Create it once with:
+#   git checkout --orphan gh-pages
+#   git reset --hard
+#   git commit --allow-empty -m "init gh-pages"
+#   git push origin gh-pages
+#
+# The benchmark dashboard will be available at:
+#   https://<org>.github.io/<repo>/dev/bench/
+name: Benchmarks
+
+on:
+  push:
+    branches: ["main"]
+  pull_request:
+    branches: ["main"]
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{ github.event_name == 'pull_request' }}
+
+permissions:
+  contents: write
+  deployments: write
+  pull-requests: write
+
+env:
+  CARGO_TERM_COLOR: always
+  CARGO_INCREMENTAL: 0
+  CARGO_NET_RETRY: 10
+  RUSTUP_MAX_RETRIES: 10
+
+jobs:
+  benchmark:
+    name: Continuous benchmarking
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Install Rust toolchain
+        uses: dtolnay/rust-toolchain@efa25f7f19611383d5b0ccf2d1c8914531636bf9 # stable
+        id: rust-toolchain
+        with:
+          toolchain: stable
+
+      - uses: actions/cache@v4
+        with:
+          path: |
+            ~/.cargo/bin/
+            ~/.cargo/registry/index/
+            ~/.cargo/registry/cache/
+            ~/.cargo/git/db/
+            target/
+          key: ${{ runner.os }}-bench-${{ steps.rust-toolchain.outputs.cachekey }}-${{ hashFiles('**/Cargo.lock') }}
+          restore-keys: |
+            ${{ runner.os }}-bench-${{ steps.rust-toolchain.outputs.cachekey }}-
+            ${{ runner.os }}-bench-
+
+      # TODO(template) update the bench command if you use a different
+      # benchmark harness or need to pass extra flags
+      - name: Run benchmarks
+        shell: bash
+        run: |
+          set -euo pipefail
+          cargo bench -p template_crate --bench benchmark -- --output-format bencher 2>/dev/null | tee benchmark-output.txt
+
+      # On PRs: restore the most-recent baseline written by a main-branch run.
+      # Using only restore-keys (no matching primary key) guarantees we always
+      # get the latest entry; the primary key includes the run ID so each main
+      # run saves a fresh entry and PRs never hit a stale one.
+      # Note: if no baseline cache exists yet (e.g. the very first PR before any
+      # main-branch run has completed), the restore will miss and no regression
+      # alert will fire. This is expected and benign — the action creates a fresh
+      # data file and the next main-branch run will establish the baseline.
+      - name: Restore benchmark baseline
+        if: github.event_name == 'pull_request'
+        uses: actions/cache/restore@v4
+        with:
+          path: ./cache
+          key: ${{ runner.os }}-benchmark-baseline-${{ github.run_id }}
+          restore-keys: |
+            ${{ runner.os }}-benchmark-baseline-
+
+      - name: Store benchmark result (PR)
+        if: github.event_name == 'pull_request'
+        uses: benchmark-action/github-action-benchmark@4bdcce38c94cec68da58d012ac24b7b1155efe8b # v1
+        with:
+          # TODO(template) update the name to match your project
+          name: Rust Template Benchmark
+          tool: "cargo"
+          output-file-path: benchmark-output.txt
+          external-data-json-path: ./cache/benchmark-data.json
+          alert-threshold: "130%"
+          comment-on-alert: true
+          fail-on-alert: true
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+
+      # On main: push results to gh-pages for the historical dashboard, then
+      # save the updated baseline so future PRs compare against current main.
+      # external-data-json-path writes the latest result into ./cache so the
+      # subsequent save step has data to cache for PR comparisons.
+      - name: Store benchmark result (main)
+        if: github.event_name == 'push'
+        uses: benchmark-action/github-action-benchmark@4bdcce38c94cec68da58d012ac24b7b1155efe8b # v1
+        with:
+          # TODO(template) update the name to match your project
+          name: Rust Template Benchmark
+          tool: "cargo"
+          output-file-path: benchmark-output.txt
+          external-data-json-path: ./cache/benchmark-data.json
+          gh-pages-branch: gh-pages
+          benchmark-data-dir-path: dev/bench
+          alert-threshold: "130%"
+          comment-on-alert: true
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          auto-push: true
+
+      # Save the freshly-updated benchmark data as the new baseline.
+      # The key embeds the run ID so every main run creates a new cache entry;
+      # actions/cache would skip saving if the key already existed.
+      - name: Save benchmark baseline
+        if: github.event_name == 'push'
+        uses: actions/cache/save@v4
+        with:
+          path: ./cache
+          key: ${{ runner.os }}-benchmark-baseline-${{ github.run_id }}
diff --git a/.github/workflows/build-and-test.yml b/.github/workflows/build-and-test.yml
index 9c3779f..2d69a8c 100644
--- a/.github/workflows/build-and-test.yml
+++ b/.github/workflows/build-and-test.yml
@@ -38,5 +38,7 @@ jobs:
       - uses: Swatinem/rust-cache@v2
       - name: Build
         run: cargo test --workspace --verbose --no-run
+      - name: Build all targets
+        run: cargo build --workspace --all-features --all-targets
       - name: Run tests
-        run: cargo test --workspace --verbose
+        run: cargo test --workspace --verbose
\ No newline at end of file
diff --git a/template_crate/Cargo.toml b/template_crate/Cargo.toml
index e0b327d..57c4ab9 100644
--- a/template_crate/Cargo.toml
+++ b/template_crate/Cargo.toml
@@ -15,10 +15,15 @@ license.workspace = true
 thiserror = { version = "1.0", default-features = false }
 
 [dev-dependencies]
+criterion = { version = "0.5", features = ["html_reports"] }
 proptest = { workspace = true }
 
 # TODO(template)
 # don't forget to put this at every crate
 # to inherit workspace's lints
+[[bench]]
+name = "benchmark"
+harness = false
+
 [lints]
 workspace = true
diff --git a/template_crate/benches/benchmark.rs b/template_crate/benches/benchmark.rs
new file mode 100644
index 0000000..804ff84
--- /dev/null
+++ b/template_crate/benches/benchmark.rs
@@ -0,0 +1,32 @@
+//! Benchmarks for template_crate
+
+use criterion::{Criterion, black_box, criterion_group};
+use template_crate::{add_small_integers, sub_small_integers};
+
+/// Benchmarks for `add_small_integers` function
+fn bench_add_small_integers(c: &mut Criterion) {
+    c.bench_function("add_small_integers valid", |b| {
+        b.iter(|| add_small_integers(black_box(50), black_box(30)))
+    });
+
+    c.bench_function("add_small_integers bound check", |b| {
+        b.iter(|| add_small_integers(black_box(200), black_box(5)))
+    });
+}
+
+/// Benchmarks for `sub_small_integers` function
+fn bench_sub_small_integers(c: &mut Criterion) {
+    c.bench_function("sub_small_integers valid", |b| {
+        b.iter(|| sub_small_integers(black_box(50), black_box(30)))
+    });
+
+    c.bench_function("sub_small_integers bound check", |b| {
+        b.iter(|| sub_small_integers(black_box(200), black_box(5)))
+    });
+}
+
+/// Entry point for benchmarks
+fn main() {
+    criterion_group!(benches, bench_add_small_integers, bench_sub_small_integers);
+    benches();
+}