diff --git a/.github/workflows/codspeed.yml b/.github/workflows/codspeed.yml
index 67fa10d..c21e07b 100644
--- a/.github/workflows/codspeed.yml
+++ b/.github/workflows/codspeed.yml
@@ -8,23 +8,23 @@ on:
 
 env:
   PYTHON_VERSION: "3.12"
+  SHARDS: 4
 
 jobs:
-  benchmarks-instrumentation:
+  benchmarks:
     strategy:
       matrix:
-        include:
-          - mode: "instrumentation"
-            runs-on: ubuntu-24.04
-          - mode: "walltime"
-            runs-on: codspeed-macro
+        shard: [1, 2, 3, 4]
+        mode: ["instrumentation", "walltime"]
 
-    name: Run ${{ matrix.mode }} benchmarks
-    runs-on: ${{ matrix.runs-on }}
+    name: "Run ${{ matrix.mode }} benchmarks (Shard #${{ matrix.shard }})"
+    runs-on: ${{ matrix.mode == 'instrumentation' && 'ubuntu-24.04' || 'codspeed-macro' }}
     steps:
       - uses: actions/checkout@v4
         with:
           submodules: "recursive"
+      - name: Install required-version defined in uv.toml
+        uses: astral-sh/setup-uv@v5
       - uses: actions/setup-python@v2
         with:
           python-version: ${{ env.PYTHON_VERSION }}
@@ -32,10 +32,18 @@ jobs:
         run: |
           sudo apt-get update
           sudo apt-get install valgrind -y
-          pip install .
+          uv sync --dev
           sudo apt-get remove valgrind -y
       - name: Run benchmarks
         uses: CodSpeedHQ/action@main
         with:
-          run: pytest tests/benchmarks/ --codspeed
+          mode: ${{ matrix.mode }}
+          run: uv run pytest tests/benchmarks/ --codspeed --test-group=${{ matrix.shard }} --test-group-count=${{ env.SHARDS }}
           token: ${{ secrets.CODSPEED_TOKEN }}
+
+  all-checks:
+    runs-on: ubuntu-latest
+    steps:
+      - run: echo "All CI checks passed."
+    needs:
+      - benchmarks
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 188f08e..2d580e3 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -5,26 +5,207 @@
 
 
 
+## [4.1.0] - 2025-10-06
+
+### <!-- 1 -->🐛 Bug Fixes
+- Use ParamSpec in instrument definition instead of tuple and dict by @art049
+- Use ParamSpec in the plugin call definition by @flying-sheep
+
+### <!-- 3 -->📚 Documentation
+- Add pull_request trigger comment by @adriencaccia
+
+### <!-- 7 -->⚙️ Internals
+- Exclude beta tags in changelog by @art049
+- Bump instrument-hooks (#87) by @not-matthias in [#87](https://github.com/CodSpeedHQ/pytest-codspeed/pull/87)
+- Switch to sharded upload (#75) by @art049 in [#75](https://github.com/CodSpeedHQ/pytest-codspeed/pull/75)
+- Fix changelog generation by @art049
+
+
+# Changelog
+
+<sub>The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).</sub>
+
+## [Unreleased]
+
+### <!-- 7 -->⚙️ Internals
+
+- Fix changelog generation by @art049
+
 ## [4.0.0] - 2025-07-10
 
 ### <!-- 0 -->🚀 Features
+
 - Update readme by @art049
+
+### <!-- 7 -->⚙️ Internals
+
+- Remove pre-releases from git-cliff changelog by @art049
+- Link to the documentation by @art049
+- Improve reliability of perf trampoline compatibility checks by @art049
+
+## [4.0.0-beta1] - 2025-06-10
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Reenable walltime instrument hooks by @art049 in [#82](https://github.com/CodSpeedHQ/pytest-codspeed/pull/82)
+
+## [4.0.0-beta] - 2025-06-06
+
+### <!-- 0 -->🚀 Features
+
 - Support pytest-benchmark's pedantic API by @art049 in [#81](https://github.com/CodSpeedHQ/pytest-codspeed/pull/81)
 - Make sure the benchmark fixture can only be called once per bench by @art049
 - Support marker attributes to customize the walltime execution by @art049 in [#80](https://github.com/CodSpeedHQ/pytest-codspeed/pull/80)
 - Use instrument hooks by @not-matthias
 - Add instrument-hooks native module by @not-matthias
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Fix native library typing by @art049
+
+### <!-- 6 -->🧪 Testing
+
+- Add benches from the documentation's getting started by @art049 in [#71](https://github.com/CodSpeedHQ/pytest-codspeed/pull/71)
+- Add simple python benches by @art049
+
+### <!-- 7 -->⚙️ Internals
+
+- Bump ruff by @art049
+- Update release workflow to include submodules by @art049 in [#79](https://github.com/CodSpeedHQ/pytest-codspeed/pull/79)
+- Remove valgrind wrapper by @not-matthias
+- Update apt before installing packages by @art049
+
+## [3.2.0] - 2025-01-31
+
+### <!-- 0 -->🚀 Features
+
 - Increase the min round time to a bigger value (+/- 1ms) by @art049
 - Add benchmarks-walltime job to run additional performance benchmarks by @art049 in [#65](https://github.com/CodSpeedHQ/pytest-codspeed/pull/65)
 - Fix the random seed while measuring with instruments by @art049 in [#48](https://github.com/CodSpeedHQ/pytest-codspeed/pull/48)
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Use time per iteration instead of total round time in stats by @art049
+
+### <!-- 2 -->🏗️ Refactor
+
+- Replace hardcoded outlier factor for improved readability by @art049 in [#67](https://github.com/CodSpeedHQ/pytest-codspeed/pull/67)
+
+### <!-- 7 -->⚙️ Internals
+
+- Fix self-dependency by @adriencaccia in [#66](https://github.com/CodSpeedHQ/pytest-codspeed/pull/66)
+- Fix uv version in CI by @adriencaccia
+
+## [3.1.2] - 2025-01-09
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Update package_data to include header and source files for valgrind wrapper by @art049 in [#64](https://github.com/CodSpeedHQ/pytest-codspeed/pull/64)
+
+## [3.1.1] - 2025-01-07
+
+### <!-- 7 -->⚙️ Internals
+
+- Fix tag num with bumpver by @art049 in [#61](https://github.com/CodSpeedHQ/pytest-codspeed/pull/61)
+- Update uv lock before release by @art049
+- Add a py3-none-any fallback wheel by @art049
+
+## [3.1.0] - 2024-12-09
+
+### <!-- 2 -->🏗️ Refactor
+
+- Remove the scripted semver generation by @art049
+
+### <!-- 7 -->⚙️ Internals
+
+- Fix typo in cibuildwheel config by @art049 in [#57](https://github.com/CodSpeedHQ/pytest-codspeed/pull/57)
+
+## [3.1.0-beta] - 2024-12-06
+
+### <!-- 0 -->🚀 Features
+
 - Check buildability and fallback when build doesn't work by @art049
 - Compile the callgrind wrapper at build time by @art049
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Allow build on arm64 by @art049
+
+### <!-- 7 -->⚙️ Internals
+
+- Build wheels with cibuildwheel by @art049
+- Allow forcing integrated tests by @art049
+- Fix release script by @art049
+- Use bumpver to manage versions by @art049
+- Add a changelog by @art049
+- Force native extension build in CI by @art049
+- Updated matrix release workflow by @art049
+- Use a common python version in the codspeed job by @art049
+- Fix the codspeed workflow by @art049
+- Use uv in CI by @art049
+- Commit uv lock file by @art049
+
+## [3.0.0] - 2024-10-29
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Fix compatibility with pytest-benchmark 5.0.0 by @art049 in [#54](https://github.com/CodSpeedHQ/pytest-codspeed/pull/54)
+
+### <!-- 7 -->⚙️ Internals
+
+- Drop support for python3.8 by @art049
+- Expose type information (#53) by @Dreamsorcerer in [#53](https://github.com/CodSpeedHQ/pytest-codspeed/pull/53)
+- Run the CI with ubuntu 24.04 by @art049
+- Improve naming in workflow examples by @art049
+- Bump actions/checkout to v4 (#47) by @fargito in [#47](https://github.com/CodSpeedHQ/pytest-codspeed/pull/47)
+
+## [3.0.0b4] - 2024-09-27
+
+### <!-- 0 -->🚀 Features
+
 - Send more outlier data by @art049
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Fix display of parametrized tests by @art049
+- Reenable gc logic by @art049
+
+### <!-- 6 -->🧪 Testing
+
+- Add benches for various syscalls by @art049
+
+## [3.0.0b3] - 2024-09-26
+
+### <!-- 0 -->🚀 Features
+
 - Also save the lower and upper fences in the json data by @art049 in [#46](https://github.com/CodSpeedHQ/pytest-codspeed/pull/46)
+
+### <!-- 6 -->🧪 Testing
+
+- Refactor the algorithm benches using parametrization and add benches on bit_manipulation by @art049
+
+## [3.0.0b2] - 2024-09-24
+
+### <!-- 0 -->🚀 Features
+
 - Also save the q1 and q3 in the json data by @art049 in [#45](https://github.com/CodSpeedHQ/pytest-codspeed/pull/45)
 - Add the --codspeed-max-time flag by @art049
+
+## [3.0.0b1] - 2024-09-20
+
+### <!-- 0 -->🚀 Features
+
 - Send the semver version to cospeed instead of the PEP440 one by @art049 in [#44](https://github.com/CodSpeedHQ/pytest-codspeed/pull/44)
 - Also store the semver version by @art049
+
+### <!-- 6 -->🧪 Testing
+
+- Add benches for TheAlgorithms/backtracking by @art049 in [#43](https://github.com/CodSpeedHQ/pytest-codspeed/pull/43)
+
+## [3.0.0b0] - 2024-09-18
+
+### <!-- 0 -->🚀 Features
+
 - Improve table style when displaying results by @art049 in [#41](https://github.com/CodSpeedHQ/pytest-codspeed/pull/41)
 - Add the total bench time to the collected stats by @art049
 - Add configuration and split tests between instruments by @art049
@@ -32,110 +213,107 @@
 - Implement the walltime instrument by @art049
 - Add bench of various python noop by @art049
 - Avoid overriding pytest's default protocol (#32) by @kenodegard in [#32](https://github.com/CodSpeedHQ/pytest-codspeed/pull/32)
-- Support pytest 8.1.1 by @art049
-- Avoid concurrent wrapper builds by @art049
-- Add a test for pytest-xdist compatibility by @art049
-- Release the package from the CI with trusted provider by @art049
-- Add a return type to the benchmark fixture by @art049 in [#13](https://github.com/CodSpeedHQ/pytest-codspeed/pull/13)
-- Add support for returning values (#12) by @patrick91 in [#12](https://github.com/CodSpeedHQ/pytest-codspeed/pull/12)
-- Warmup performance map generation by @art049
-- Add some details about the callgraph generation status in the header by @art049
-- Test that perf maps are generated by @art049
-- Add a local test matrix with hatch by @art049
-- Test that benchmark selection with -k works by @art049
-- Add support for CPython3.12 and perf trampoline by @art049
-- Add introspection benchmarks by @art049 in [#9](https://github.com/CodSpeedHQ/pytest-codspeed/pull/9)
-- Add library metadata in the profile output by @art049 in [#5](https://github.com/CodSpeedHQ/pytest-codspeed/pull/5)
-- Allow running along with pytest-benchmarks by @art049
 
 ### <!-- 1 -->🐛 Bug Fixes
-- Reenable walltime instrument hooks by @art049 in [#82](https://github.com/CodSpeedHQ/pytest-codspeed/pull/82)
-- Fix native library typing by @art049
-- Use time per iteration instead of total round time in stats by @art049
-- Update package_data to include header and source files for valgrind wrapper by @art049 in [#64](https://github.com/CodSpeedHQ/pytest-codspeed/pull/64)
-- Allow build on arm64 by @art049
-- Fix compatibility with pytest-benchmark 5.0.0 by @art049 in [#54](https://github.com/CodSpeedHQ/pytest-codspeed/pull/54)
-- Fix display of parametrized tests by @art049
-- Reenable gc logic by @art049
+
 - Use importlib_metadata to keep backward compatibility by @art049
 - Properly decide the mode depending on our env variable spec by @art049
 - Disable pytest-speed when installed and codspeed is enabled by @art049
-- Loosen runtime requirements (#21) by @edgarrmondragon in [#21](https://github.com/CodSpeedHQ/pytest-codspeed/pull/21)
-- Fix xdist test output assertion by @art049
-- Fix relative git path when using working-directory by @art049 in [#15](https://github.com/CodSpeedHQ/pytest-codspeed/pull/15)
-- Fix typo in release.yml (#14) by @art049 in [#14](https://github.com/CodSpeedHQ/pytest-codspeed/pull/14)
-- Fix setuptools installation with python3.12 by @art049
-- Support benchmark.extra_info parameters on the fixture by @art049 in [#10](https://github.com/CodSpeedHQ/pytest-codspeed/pull/10)
-- Filter out pytest-benchmark warnings in the tests by @art049
-- Support kwargs with the benchmark fixture by @art049 in [#4](https://github.com/CodSpeedHQ/pytest-codspeed/pull/4)
-- Avoid wrapping the callable to maintain existing results by @art049
-- Disable automatic garbage collection to increase stability by @art049 in [#2](https://github.com/CodSpeedHQ/pytest-codspeed/pull/2)
-- Update readme by @art049
-- Fix the release script by @art049
-- Make the release script executable by @art049
-- Match the test output in any order by @art049
 
 ### <!-- 2 -->🏗️ Refactor
-- Replace hardcoded outlier factor for improved readability by @art049 in [#67](https://github.com/CodSpeedHQ/pytest-codspeed/pull/67)
-- Remove the scripted semver generation by @art049
+
 - Differentiate the mode from the underlying instrument by @art049
 - Move the instrumentation wrapper directly in the instrument by @art049
 - Change Instrumentation to CPUInstrumentation by @art049
 - Create an abstraction for each instrument by @art049
-- Use the pytest_run_protocol hook for better exec control by @art049
-- Manage compatibility env in the conftest by @art049
 
 ### <!-- 3 -->📚 Documentation
+
 - Update action version in the CI workflow configuration (#39) by @frgfm in [#39](https://github.com/CodSpeedHQ/pytest-codspeed/pull/39)
 - Bump action versions in README by @adriencaccia
 
 ### <!-- 6 -->🧪 Testing
-- Add benches from the documentation's getting started by @art049 in [#71](https://github.com/CodSpeedHQ/pytest-codspeed/pull/71)
-- Add simple python benches by @art049
-- Add benches for various syscalls by @art049
-- Refactor the algorithm benches using parametrization and add benches on bit_manipulation by @art049
-- Add benches for TheAlgorithms/backtracking by @art049 in [#43](https://github.com/CodSpeedHQ/pytest-codspeed/pull/43)
+
 - Add benches for TheAlgorithms/audio_filters by @art049 in [#42](https://github.com/CodSpeedHQ/pytest-codspeed/pull/42)
 
 ### <!-- 7 -->⚙️ Internals
-- Remove pre-releases from git-cliff changelog
-- Link to the documentation by @art049
-- Improve reliability of perf trampoline compatibility checks by @art049
-- Bump ruff by @art049
-- Update release workflow to include submodules by @art049 in [#79](https://github.com/CodSpeedHQ/pytest-codspeed/pull/79)
-- Remove valgrind wrapper by @not-matthias
-- Update apt before installing packages by @art049
-- Fix self-dependency by @adriencaccia in [#66](https://github.com/CodSpeedHQ/pytest-codspeed/pull/66)
-- Fix uv version in CI by @adriencaccia
-- Fix tag num with bumpver by @art049 in [#61](https://github.com/CodSpeedHQ/pytest-codspeed/pull/61)
-- Update uv lock before release by @art049
-- Add a py3-none-any fallback wheel by @art049
-- Fix typo in cibuildwheel config by @art049 in [#57](https://github.com/CodSpeedHQ/pytest-codspeed/pull/57)
-- Build wheels with cibuildwheel by @art049
-- Allow forcing integrated tests by @art049
-- Fix release script by @art049
-- Use bumpver to manage versions by @art049
-- Add a changelog by @art049
-- Force native extension build in CI by @art049
-- Updated matrix release workflow by @art049
-- Use a common python version in the codspeed job by @art049
-- Fix the codspeed workflow by @art049
-- Use uv in CI by @art049
-- Commit uv lock file by @art049
-- Drop support for python3.8 by @art049
-- Expose type information (#53) by @Dreamsorcerer in [#53](https://github.com/CodSpeedHQ/pytest-codspeed/pull/53)
-- Run the CI with ubuntu 24.04 by @art049
-- Improve naming in workflow examples by @art049
-- Bump actions/checkout to v4 (#47) by @fargito in [#47](https://github.com/CodSpeedHQ/pytest-codspeed/pull/47)
+
 - Add a test on the walltime instrument by @art049
 - Fix utils test using a fake git repo by @art049
 - Update readme by @art049
 - Support python 3.13 and drop 3.7 by @art049 in [#40](https://github.com/CodSpeedHQ/pytest-codspeed/pull/40)
 - Add TCH, FA, and UP to ruff lints (#29) by @kenodegard in [#29](https://github.com/CodSpeedHQ/pytest-codspeed/pull/29)
+
+## [2.2.1] - 2024-03-19
+
+### <!-- 0 -->🚀 Features
+
+- Support pytest 8.1.1 by @art049
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Loosen runtime requirements (#21) by @edgarrmondragon in [#21](https://github.com/CodSpeedHQ/pytest-codspeed/pull/21)
+
+### <!-- 7 -->⚙️ Internals
+
 - Add all-checks job to CI workflow by @art049 in [#28](https://github.com/CodSpeedHQ/pytest-codspeed/pull/28)
 - Switch from black to ruff format by @art049
 - Update action version in README.md by @adriencaccia
 - Add codspeed badge to the readme by @art049
+
+## [2.2.0] - 2023-09-01
+
+### <!-- 0 -->🚀 Features
+
+- Avoid concurrent wrapper builds by @art049
+- Add a test for pytest-xdist compatibility by @art049
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Fix xdist test output assertion by @art049
+
+## [2.1.0] - 2023-07-27
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Fix relative git path when using working-directory by @art049 in [#15](https://github.com/CodSpeedHQ/pytest-codspeed/pull/15)
+- Fix typo in release.yml (#14) by @art049 in [#14](https://github.com/CodSpeedHQ/pytest-codspeed/pull/14)
+
+## [2.0.1] - 2023-07-22
+
+### <!-- 0 -->🚀 Features
+
+- Release the package from the CI with trusted provider by @art049
+- Add a return type to the benchmark fixture by @art049 in [#13](https://github.com/CodSpeedHQ/pytest-codspeed/pull/13)
+- Add support for returning values (#12) by @patrick91 in [#12](https://github.com/CodSpeedHQ/pytest-codspeed/pull/12)
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Fix setuptools installation with python3.12 by @art049
+
+## [2.0.0] - 2023-07-04
+
+### <!-- 0 -->🚀 Features
+
+- Warmup performance map generation by @art049
+- Add some details about the callgraph generation status in the header by @art049
+- Test that perf maps are generated by @art049
+- Add a local test matrix with hatch by @art049
+- Test that benchmark selection with -k works by @art049
+- Add support for CPython3.12 and perf trampoline by @art049
+- Add introspection benchmarks by @art049 in [#9](https://github.com/CodSpeedHQ/pytest-codspeed/pull/9)
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Support benchmark.extra_info parameters on the fixture by @art049 in [#10](https://github.com/CodSpeedHQ/pytest-codspeed/pull/10)
+- Filter out pytest-benchmark warnings in the tests by @art049
+
+### <!-- 2 -->🏗️ Refactor
+
+- Use the pytest_run_protocol hook for better exec control by @art049
+
+### <!-- 7 -->⚙️ Internals
+
 - Separate the benchmark workflow by @art049 in [#8](https://github.com/CodSpeedHQ/pytest-codspeed/pull/8)
 - Bump version to 1.3.0 to trigger the callgraph generation by @art049
 - Reuse same test code in the tests by @art049
@@ -145,12 +323,79 @@
 - Restructure dev dependencies by @art049
 - Replace isort by ruff by @art049 in [#11](https://github.com/CodSpeedHQ/pytest-codspeed/pull/11)
 - Add discord badge in the readme by @art049
+
+## [1.2.2] - 2022-12-02
+
+### <!-- 0 -->🚀 Features
+
+- Add library metadata in the profile output by @art049 in [#5](https://github.com/CodSpeedHQ/pytest-codspeed/pull/5)
+
+## [1.2.1] - 2022-11-28
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Support kwargs with the benchmark fixture by @art049 in [#4](https://github.com/CodSpeedHQ/pytest-codspeed/pull/4)
+
+## [1.2.0] - 2022-11-22
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Avoid wrapping the callable to maintain existing results by @art049
+- Disable automatic garbage collection to increase stability by @art049 in [#2](https://github.com/CodSpeedHQ/pytest-codspeed/pull/2)
+- Update readme by @art049
+
+### <!-- 7 -->⚙️ Internals
+
 - Update readme by @art049
+
+## [1.1.0] - 2022-11-10
+
+### <!-- 0 -->🚀 Features
+
+- Allow running along with pytest-benchmarks by @art049
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- Fix the release script by @art049
+- Make the release script executable by @art049
+- Match the test output in any order by @art049
+
+### <!-- 2 -->🏗️ Refactor
+
+- Manage compatibility env in the conftest by @art049
+
+### <!-- 7 -->⚙️ Internals
+
 - Add a pytest-benchmark compatibility test by @art049 in [#1](https://github.com/CodSpeedHQ/pytest-codspeed/pull/1)
 - Add more details on the pytest run by @art049
 - Continue running on matrix item error by @art049
 - Add a CI configuration with pytest-benchmark installed by @art049
 
+## [1.0.1] - 2022-11-05
 
+[unreleased]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v4.0.0..HEAD
+[4.0.0]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v4.0.0-beta1..v4.0.0
+[4.0.0-beta1]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v4.0.0-beta..v4.0.0-beta1
+[4.0.0-beta]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v3.2.0..v4.0.0-beta
+[3.2.0]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v3.1.2..v3.2.0
+[3.1.2]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v3.1.1..v3.1.2
+[3.1.1]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v3.1.0..v3.1.1
+[3.1.0]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v3.1.0-beta..v3.1.0
+[3.1.0-beta]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v3.0.0..v3.1.0-beta
+[3.0.0]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v3.0.0b4..v3.0.0
+[3.0.0b4]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v3.0.0b3..v3.0.0b4
+[3.0.0b3]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v3.0.0b2..v3.0.0b3
+[3.0.0b2]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v3.0.0b1..v3.0.0b2
+[3.0.0b1]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v3.0.0b0..v3.0.0b1
+[3.0.0b0]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v2.2.1..v3.0.0b0
+[2.2.1]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v2.2.0..v2.2.1
+[2.2.0]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v2.1.0..v2.2.0
+[2.1.0]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v2.0.1..v2.1.0
+[2.0.1]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v2.0.0..v2.0.1
+[2.0.0]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v1.2.2..v2.0.0
+[1.2.2]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v1.2.1..v1.2.2
+[1.2.1]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v1.2.0..v1.2.1
+[1.2.0]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v1.1.0..v1.2.0
+[1.1.0]: https://github.com/CodSpeedHQ/pytest-codspeed/compare/v1.0.4..v1.1.0
 
 <!-- generated by git-cliff -->
diff --git a/README.md b/README.md
index d16e445..59aa1bb 100644
--- a/README.md
+++ b/README.md
@@ -92,7 +92,7 @@ on:
   push:
     branches:
       - "main" # or "master"
-  pull_request:
+  pull_request: # required to have reports on PRs
   # `workflow_dispatch` allows CodSpeed to trigger backtest
   # performance analysis in order to generate initial data.
   workflow_dispatch:
@@ -111,8 +111,9 @@ jobs:
         run: pip install -r requirements.txt
 
       - name: Run benchmarks
-        uses: CodSpeedHQ/action@v3
+        uses: CodSpeedHQ/action@v4
         with:
+          mode: instrumentation   # or `walltime`
           token: ${{ secrets.CODSPEED_TOKEN }}
           run: pytest tests/ --codspeed
 ```
diff --git a/cliff.toml b/cliff.toml
index 5e9bba6..8c0ca49 100644
--- a/cliff.toml
+++ b/cliff.toml
@@ -72,8 +72,6 @@ postprocessors = [
 # output = "test.md"
 
 [git]
-# ignore pre-release tags
-ignore_tags = "rc|alpha|beta"
 # parse the commits based on https://www.conventionalcommits.org
 conventional_commits = true
 # filter out the commits that are not conventional
@@ -88,6 +86,7 @@ commit_preprocessors = [
     # If the spelling is incorrect, it will be automatically fixed.
     #{ pattern = '.*', replace_command = 'typos --write-changes -' },
 ]
+ignore_tags = ".*beta.*"
 # regex for parsing and grouping commits
 commit_parsers = [
     { message = "^feat", group = "<!-- 0 -->🚀 Features" },
diff --git a/pyproject.toml b/pyproject.toml
index 7d1ed88..2295312 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -37,19 +37,24 @@ dependencies = [
 ]
 
 [project.optional-dependencies]
-lint = ["mypy ~= 1.11.2", "ruff ~= 0.11.12"]
 compat = [
     "pytest-benchmark ~= 5.0.0",
     "pytest-xdist ~= 3.6.1",
     # "pytest-speed>=0.3.5",
 ]
-test = ["pytest ~= 7.0", "pytest-cov ~= 4.0.0"]
 
 [tool.uv.sources]
 pytest-codspeed = { workspace = true }
 
 [dependency-groups]
-dev = ["pytest-codspeed"]
+dev = [
+    "pytest-codspeed",
+    "mypy ~= 1.11.2",
+    "ruff ~= 0.11.12",
+    "pytest ~= 7.0",
+    "pytest-cov ~= 4.0.0",
+    "pytest-test-groups>=1.1.0",
+]
 
 [project.entry-points]
 pytest11 = { codspeed = "pytest_codspeed.plugin" }
@@ -66,7 +71,7 @@ version = { attr = "pytest_codspeed.__version__" }
 
 
 [tool.bumpver]
-current_version = "4.0.0"
+current_version = "4.1.0"
 version_pattern = "MAJOR.MINOR.PATCH[-TAG[NUM]]"
 commit_message = "Release v{new_version} 🚀"
 tag_message = "Release v{new_version} 🚀"
diff --git a/scripts/pre-release.sh b/scripts/pre-release.sh
index aedfe1d..99525ad 100755
--- a/scripts/pre-release.sh
+++ b/scripts/pre-release.sh
@@ -13,7 +13,7 @@ else
         echo "GITHUB_TOKEN is not set. Trying to fetch it from gh"
         GITHUB_TOKEN=$(gh auth token)
     fi
-    git cliff -o CHANGELOG.md --tag $VERSION
+    git cliff --unreleased --tag $VERSION --prepend CHANGELOG.md
     git add CHANGELOG.md
 fi
 
diff --git a/src/pytest_codspeed/__init__.py b/src/pytest_codspeed/__init__.py
index 27e4f87..3644fe5 100644
--- a/src/pytest_codspeed/__init__.py
+++ b/src/pytest_codspeed/__init__.py
@@ -1,6 +1,6 @@
-__version__ = "4.0.0"
+__version__ = "4.1.0"
 # We also have the semver version since __version__ is not semver compliant
-__semver_version__ = "4.0.0"
+__semver_version__ = "4.1.0"
 
 from .plugin import BenchmarkFixture
 
diff --git a/src/pytest_codspeed/instruments/__init__.py b/src/pytest_codspeed/instruments/__init__.py
index fb264f1..71936db 100644
--- a/src/pytest_codspeed/instruments/__init__.py
+++ b/src/pytest_codspeed/instruments/__init__.py
@@ -8,11 +8,13 @@
     from typing import Any, Callable, ClassVar, TypeVar
 
     import pytest
+    from typing_extensions import ParamSpec
 
     from pytest_codspeed.config import BenchmarkMarkerOptions, PedanticOptions
     from pytest_codspeed.plugin import CodSpeedConfig
 
     T = TypeVar("T")
+    P = ParamSpec("P")
 
 
 class Instrument(metaclass=ABCMeta):
@@ -30,9 +32,9 @@ def measure(
         marker_options: BenchmarkMarkerOptions,
         name: str,
         uri: str,
-        fn: Callable[..., T],
-        *args: tuple,
-        **kwargs: dict[str, Any],
+        fn: Callable[P, T],
+        *args: P.args,
+        **kwargs: P.kwargs,
     ) -> T: ...
 
     @abstractmethod
diff --git a/src/pytest_codspeed/instruments/hooks/__init__.py b/src/pytest_codspeed/instruments/hooks/__init__.py
index 98e32fa..19aea91 100644
--- a/src/pytest_codspeed/instruments/hooks/__init__.py
+++ b/src/pytest_codspeed/instruments/hooks/__init__.py
@@ -63,7 +63,7 @@ def set_executed_benchmark(self, uri: str, pid: int | None = None) -> None:
         if pid is None:
             pid = os.getpid()
 
-        ret = self.lib.instrument_hooks_executed_benchmark(
+        ret = self.lib.instrument_hooks_set_executed_benchmark(
             self.instance, pid, uri.encode("ascii")
         )
         if ret != 0:
diff --git a/src/pytest_codspeed/instruments/hooks/build.py b/src/pytest_codspeed/instruments/hooks/build.py
index 99071d1..992ee01 100644
--- a/src/pytest_codspeed/instruments/hooks/build.py
+++ b/src/pytest_codspeed/instruments/hooks/build.py
@@ -6,10 +6,35 @@
 
 includes_dir = Path(__file__).parent.joinpath("instrument-hooks/includes")
 header_text = (includes_dir / "core.h").read_text()
-filtered_header = "\n".join(
-    line for line in header_text.splitlines() if not line.strip().startswith("#")
-)
-ffibuilder.cdef(filtered_header)
+
+
+# Manually copied from `instrument-hooks/includes/core.h` to avoid parsing issues
+ffibuilder.cdef("""
+typedef uint64_t *InstrumentHooks;
+
+InstrumentHooks *instrument_hooks_init(void);
+void instrument_hooks_deinit(InstrumentHooks *);
+
+bool instrument_hooks_is_instrumented(InstrumentHooks *);
+uint8_t instrument_hooks_start_benchmark(InstrumentHooks *);
+uint8_t instrument_hooks_stop_benchmark(InstrumentHooks *);
+uint8_t instrument_hooks_set_executed_benchmark(InstrumentHooks *, int32_t pid,
+                                               const char *uri);
+uint8_t instrument_hooks_set_integration(InstrumentHooks *, const char *name,
+                                        const char *version);
+
+#define MARKER_TYPE_SAMPLE_START 0
+#define MARKER_TYPE_SAMPLE_END 1
+#define MARKER_TYPE_BENCHMARK_START 2
+#define MARKER_TYPE_BENCHMARK_END 3
+
+uint8_t instrument_hooks_add_marker(InstrumentHooks *, uint32_t pid,
+                                   uint8_t marker_type, uint64_t timestamp);
+uint64_t instrument_hooks_current_timestamp(void);
+
+void callgrind_start_instrumentation();
+void callgrind_stop_instrumentation();
+""")
 
 ffibuilder.set_source(
     "pytest_codspeed.instruments.hooks.dist_instrument_hooks",
diff --git a/src/pytest_codspeed/instruments/hooks/dist_instrument_hooks.pyi b/src/pytest_codspeed/instruments/hooks/dist_instrument_hooks.pyi
index 3ae9ec5..37d1a79 100644
--- a/src/pytest_codspeed/instruments/hooks/dist_instrument_hooks.pyi
+++ b/src/pytest_codspeed/instruments/hooks/dist_instrument_hooks.pyi
@@ -12,7 +12,7 @@ class lib:
     @staticmethod
     def instrument_hooks_stop_benchmark(hooks: InstrumentHooksPointer) -> int: ...
     @staticmethod
-    def instrument_hooks_executed_benchmark(
+    def instrument_hooks_set_executed_benchmark(
         hooks: InstrumentHooksPointer, pid: int, uri: bytes
     ) -> int: ...
     @staticmethod
@@ -20,6 +20,12 @@ class lib:
         hooks: InstrumentHooksPointer, name: bytes, version: bytes
     ) -> int: ...
     @staticmethod
+    def instrument_hooks_add_marker(
+        hooks: InstrumentHooksPointer, pid: int, marker_type: int, timestamp: int
+    ) -> int: ...
+    @staticmethod
+    def instrument_hooks_current_timestamp() -> int: ...
+    @staticmethod
     def callgrind_start_instrumentation() -> int: ...
     @staticmethod
     def callgrind_stop_instrumentation() -> int: ...
diff --git a/src/pytest_codspeed/instruments/hooks/instrument-hooks b/src/pytest_codspeed/instruments/hooks/instrument-hooks
index b003e50..b1e401a 160000
--- a/src/pytest_codspeed/instruments/hooks/instrument-hooks
+++ b/src/pytest_codspeed/instruments/hooks/instrument-hooks
@@ -1 +1 @@
-Subproject commit b003e5024d61cfb784d6ac6f3ffd7d61bf7b9ec9
+Subproject commit b1e401a4d031ad308edb22ed59a52253a1ebe924
diff --git a/src/pytest_codspeed/instruments/valgrind.py b/src/pytest_codspeed/instruments/valgrind.py
index 71b1542..dcb5eea 100644
--- a/src/pytest_codspeed/instruments/valgrind.py
+++ b/src/pytest_codspeed/instruments/valgrind.py
@@ -14,7 +14,7 @@
     from pytest import Session
 
     from pytest_codspeed.config import PedanticOptions
-    from pytest_codspeed.instruments import T
+    from pytest_codspeed.instruments import P, T
     from pytest_codspeed.plugin import BenchmarkMarkerOptions, CodSpeedConfig
 
 
@@ -52,9 +52,9 @@ def measure(
         marker_options: BenchmarkMarkerOptions,
         name: str,
         uri: str,
-        fn: Callable[..., T],
-        *args: tuple,
-        **kwargs: dict[str, Any],
+        fn: Callable[P, T],
+        *args: P.args,
+        **kwargs: P.kwargs,
     ) -> T:
         self.benchmark_count += 1
 
diff --git a/src/pytest_codspeed/instruments/walltime.py b/src/pytest_codspeed/instruments/walltime.py
index 2423240..e2c5815 100644
--- a/src/pytest_codspeed/instruments/walltime.py
+++ b/src/pytest_codspeed/instruments/walltime.py
@@ -24,7 +24,7 @@
     from pytest import Session
 
     from pytest_codspeed.config import PedanticOptions
-    from pytest_codspeed.instruments import T
+    from pytest_codspeed.instruments import P, T
     from pytest_codspeed.plugin import BenchmarkMarkerOptions, CodSpeedConfig
 
 DEFAULT_WARMUP_TIME_NS = 1_000_000_000
@@ -187,9 +187,9 @@ def measure(
         marker_options: BenchmarkMarkerOptions,
         name: str,
         uri: str,
-        fn: Callable[..., T],
-        *args: tuple,
-        **kwargs: dict[str, Any],
+        fn: Callable[P, T],
+        *args: P.args,
+        **kwargs: P.kwargs,
     ) -> T:
         benchmark_config = BenchmarkConfig.from_codspeed_config_and_marker_data(
             self.config, marker_options
diff --git a/src/pytest_codspeed/plugin.py b/src/pytest_codspeed/plugin.py
index 24e0401..014d7d7 100644
--- a/src/pytest_codspeed/plugin.py
+++ b/src/pytest_codspeed/plugin.py
@@ -30,11 +30,12 @@
 from . import __version__
 
 if TYPE_CHECKING:
-    from typing import Any, Callable, TypeVar
+    from typing import Any, Callable, ParamSpec, TypeVar
 
     from pytest_codspeed.instruments import Instrument
 
     T = TypeVar("T")
+    P = ParamSpec("P")
 
 
 @pytest.hookimpl(trylast=True)
@@ -264,10 +265,10 @@ def wrap_runtest(
     plugin: CodSpeedPlugin,
     node: pytest.Item,
     config: pytest.Config,
-    fn: Callable[..., T],
-) -> Callable[..., T]:
+    fn: Callable[P, T],
+) -> Callable[P, T]:
     @functools.wraps(fn)
-    def wrapped(*args: tuple, **kwargs: dict[str, Any]) -> T:
+    def wrapped(*args: P.args, **kwargs: P.kwargs) -> T:
         return _measure(plugin, node, config, None, fn, args, kwargs)
 
     return wrapped
@@ -328,9 +329,7 @@ def __init__(self, request: pytest.FixtureRequest):
         self._plugin = get_plugin(self._config)
         self._called = False
 
-    def __call__(
-        self, target: Callable[..., T], *args: tuple, **kwargs: dict[str, Any]
-    ) -> T:
+    def __call__(self, target: Callable[P, T], *args: P.args, **kwargs: P.kwargs) -> T:
         if self._called:
             raise RuntimeError("The benchmark fixture can only be used once per test")
         self._called = True
diff --git a/uv.lock b/uv.lock
index 0a56fed..cc69240 100644
--- a/uv.lock
+++ b/uv.lock
@@ -346,36 +346,36 @@ compat = [
     { name = "pytest-benchmark" },
     { name = "pytest-xdist" },
 ]
-lint = [
-    { name = "mypy" },
-    { name = "ruff" },
-]
-test = [
-    { name = "pytest" },
-    { name = "pytest-cov" },
-]
 
 [package.dev-dependencies]
 dev = [
+    { name = "mypy" },
+    { name = "pytest" },
     { name = "pytest-codspeed" },
+    { name = "pytest-cov" },
+    { name = "pytest-test-groups" },
+    { name = "ruff" },
 ]
 
 [package.metadata]
 requires-dist = [
     { name = "cffi", specifier = ">=1.17.1" },
     { name = "importlib-metadata", marker = "python_full_version < '3.10'", specifier = ">=8.5.0" },
-    { name = "mypy", marker = "extra == 'lint'", specifier = "~=1.11.2" },
     { name = "pytest", specifier = ">=3.8" },
-    { name = "pytest", marker = "extra == 'test'", specifier = "~=7.0" },
     { name = "pytest-benchmark", marker = "extra == 'compat'", specifier = "~=5.0.0" },
-    { name = "pytest-cov", marker = "extra == 'test'", specifier = "~=4.0.0" },
     { name = "pytest-xdist", marker = "extra == 'compat'", specifier = "~=3.6.1" },
     { name = "rich", specifier = ">=13.8.1" },
-    { name = "ruff", marker = "extra == 'lint'", specifier = "~=0.11.12" },
 ]
 
 [package.metadata.requires-dev]
-dev = [{ name = "pytest-codspeed", editable = "." }]
+dev = [
+    { name = "mypy", specifier = "~=1.11.2" },
+    { name = "pytest", specifier = "~=7.0" },
+    { name = "pytest-codspeed", editable = "." },
+    { name = "pytest-cov", specifier = "~=4.0.0" },
+    { name = "pytest-test-groups", specifier = ">=1.1.0" },
+    { name = "ruff", specifier = "~=0.11.12" },
+]
 
 [[package]]
 name = "pytest-cov"
@@ -390,6 +390,18 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/fe/1f/9ec0ddd33bd2b37d6ec50bb39155bca4fe7085fa78b3b434c05459a860e3/pytest_cov-4.0.0-py3-none-any.whl", hash = "sha256:2feb1b751d66a8bd934e5edfa2e961d11309dc37b73b0eabe73b5945fee20f6b", size = 21554 },
 ]
 
+[[package]]
+name = "pytest-test-groups"
+version = "1.2.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pytest" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/9f/5a/c7874fe15e03d86a1109a3274b57a2473edb8a1dda4a4d27f25d848b6ff5/pytest_test_groups-1.2.1.tar.gz", hash = "sha256:67576b295522fc144b3a42fa1801f50ae962389e984b48bab4336686d09032f1", size = 8137 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/31/ff/7ff0ca5e8051931bf7fb65e31f085f7c0577615bf3a4776fb583cb471800/pytest_test_groups-1.2.1-py3-none-any.whl", hash = "sha256:8c7a016448f9ad347fb69a62f417f0a2358ecbf129fe44bc44ee991918a0bb73", size = 5278 },
+]
+
 [[package]]
 name = "pytest-xdist"
 version = "3.6.1"