diff --git a/.dockerignore b/.dockerignore
index 4f074d44..c0d8a84b 100644
--- a/.dockerignore
+++ b/.dockerignore
@@ -73,7 +73,7 @@ pip-log.txt
 pip-delete-this-directory.txt
 
 # Unit test / coverage reports
-!tests/data
+!tests/artifacts
 htmlcov/
 .tox/
 .nox/
diff --git a/.gitignore b/.gitignore
index da4b1089..d6c51c90 100644
--- a/.gitignore
+++ b/.gitignore
@@ -78,7 +78,7 @@ pip-log.txt
 pip-delete-this-directory.txt
 
 # Unit test / coverage reports
-!tests/data
+!tests/artifacts
 htmlcov/
 .tox/
 .nox/
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 21016efa..e699f543 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -12,10 +12,17 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-exclude: ^(tests/data)
+exclude: "tests/artifacts/.*\\.safetensors$"
 default_language_version:
     python: python3.10
 repos:
+  ##### Meta #####
+  - repo: meta
+    hooks:
+      - id: check-useless-excludes
+      - id: check-hooks-apply
+
+
   ##### Style / Misc. #####
   - repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v5.0.0
@@ -28,31 +35,37 @@ repos:
       - id: check-toml
       - id: end-of-file-fixer
       - id: trailing-whitespace
+
   - repo: https://github.com/crate-ci/typos
-    rev: v1.30.0
+    rev: v1.30.2
     hooks:
       - id: typos
         args: [--force-exclude]
+
   - repo: https://github.com/asottile/pyupgrade
     rev: v3.19.1
     hooks:
     -   id: pyupgrade
+
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.9.9
+    rev: v0.9.10
     hooks:
       - id: ruff
         args: [--fix]
       - id: ruff-format
 
+
   ##### Security #####
   - repo: https://github.com/gitleaks/gitleaks
     rev: v8.24.0
     hooks:
       - id: gitleaks
+
   - repo: https://github.com/woodruffw/zizmor-pre-commit
     rev: v1.4.1
     hooks:
       - id: zizmor
+
   - repo: https://github.com/PyCQA/bandit
     rev: 1.8.3
     hooks:
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index 8aff26d8..a9e4a856 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -291,7 +291,7 @@ sudo apt-get install git-lfs
 git lfs install
 ```
 
-Pull artifacts if they're not in [tests/data](tests/data)
+Pull artifacts if they're not in [tests/artifacts](tests/artifacts)
 ```bash
 git lfs pull
 ```
diff --git a/README.md b/README.md
index b16e3469..effbb08b 100644
--- a/README.md
+++ b/README.md
@@ -232,8 +232,8 @@ python lerobot/scripts/eval.py \
     --env.type=pusht \
     --eval.batch_size=10 \
     --eval.n_episodes=10 \
-    --use_amp=false \
-    --device=cuda
+    --policy.use_amp=false \
+    --policy.device=cuda
 ```
 
 Note: After training your own policy, you can re-evaluate the checkpoints with:
diff --git a/examples/4_train_policy_with_script.md b/examples/4_train_policy_with_script.md
index 58ed239a..b23d2271 100644
--- a/examples/4_train_policy_with_script.md
+++ b/examples/4_train_policy_with_script.md
@@ -1,5 +1,5 @@
 This tutorial will explain the training script, how to use it, and particularly how to configure everything needed for the training run.
-> **Note:** The following assume you're running these commands on a machine equipped with a cuda GPU. If you don't have one (or if you're using a Mac), you can add `--device=cpu` (`--device=mps` respectively). However, be advised that the code executes much slower on cpu.
+> **Note:** The following assume you're running these commands on a machine equipped with a cuda GPU. If you don't have one (or if you're using a Mac), you can add `--policy.device=cpu` (`--policy.device=mps` respectively). However, be advised that the code executes much slower on cpu.
 
 
 ## The training script
diff --git a/examples/7_get_started_with_real_robot.md b/examples/7_get_started_with_real_robot.md
index f4bef69c..894c44a3 100644
--- a/examples/7_get_started_with_real_robot.md
+++ b/examples/7_get_started_with_real_robot.md
@@ -386,14 +386,14 @@ When you connect your robot for the first time, the [`ManipulatorRobot`](../lero
 
 Here are the positions you'll move the follower arm to:
 
-| 1. Zero position | 2. Rotated position | 3. Rest position |
-|---|---|---|
+| 1. Zero position                                                                                                                                                  | 2. Rotated position                                                                                                                                                        | 3. Rest position                                                                                                                                                  |
+| ----------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | <img src="../media/koch/follower_zero.webp?raw=true" alt="Koch v1.1 follower arm zero position" title="Koch v1.1 follower arm zero position" style="width:100%;"> | <img src="../media/koch/follower_rotated.webp?raw=true" alt="Koch v1.1 follower arm rotated position" title="Koch v1.1 follower arm rotated position" style="width:100%;"> | <img src="../media/koch/follower_rest.webp?raw=true" alt="Koch v1.1 follower arm rest position" title="Koch v1.1 follower arm rest position" style="width:100%;"> |
 
 And here are the corresponding positions for the leader arm:
 
-| 1. Zero position | 2. Rotated position | 3. Rest position |
-|---|---|---|
+| 1. Zero position                                                                                                                                            | 2. Rotated position                                                                                                                                                  | 3. Rest position                                                                                                                                            |
+| ----------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | <img src="../media/koch/leader_zero.webp?raw=true" alt="Koch v1.1 leader arm zero position" title="Koch v1.1 leader arm zero position" style="width:100%;"> | <img src="../media/koch/leader_rotated.webp?raw=true" alt="Koch v1.1 leader arm rotated position" title="Koch v1.1 leader arm rotated position" style="width:100%;"> | <img src="../media/koch/leader_rest.webp?raw=true" alt="Koch v1.1 leader arm rest position" title="Koch v1.1 leader arm rest position" style="width:100%;"> |
 
 You can watch a [video tutorial of the calibration procedure](https://youtu.be/8drnU9uRY24) for more details.
@@ -898,14 +898,14 @@ python lerobot/scripts/train.py \
   --policy.type=act \
   --output_dir=outputs/train/act_koch_test \
   --job_name=act_koch_test \
-  --device=cuda \
+  --policy.device=cuda \
   --wandb.enable=true
 ```
 
 Let's explain it:
 1. We provided the dataset as argument with `--dataset.repo_id=${HF_USER}/koch_test`.
 2. We provided the policy with `policy.type=act`. This loads configurations from [`configuration_act.py`](../lerobot/common/policies/act/configuration_act.py). Importantly, this policy will automatically adapt to the number of motor sates, motor actions and cameras of your robot (e.g. `laptop` and `phone`) which have been saved in your dataset.
-4. We provided `device=cuda` since we are training on a Nvidia GPU, but you could use `device=mps` to train on Apple silicon.
+4. We provided `policy.device=cuda` since we are training on a Nvidia GPU, but you could use `policy.device=mps` to train on Apple silicon.
 5. We provided `wandb.enable=true` to use [Weights and Biases](https://docs.wandb.ai/quickstart) for visualizing training plots. This is optional but if you use it, make sure you are logged in by running `wandb login`.
 
 For more information on the `train` script see the previous tutorial: [`examples/4_train_policy_with_script.md`](../examples/4_train_policy_with_script.md)
diff --git a/lerobot/common/cameras/intel/camera_realsense.py b/lerobot/common/cameras/intel/camera_realsense.py
index 13080904..910d51c6 100644
--- a/lerobot/common/cameras/intel/camera_realsense.py
+++ b/lerobot/common/cameras/intel/camera_realsense.py
@@ -49,7 +49,7 @@ def find_cameras(raise_when_empty=True, mock=False) -> list[dict]:
     connected to the computer.
     """
     if mock:
-        import tests.mock_pyrealsense2 as rs
+        import tests.cameras.mock_pyrealsense2 as rs
     else:
         import pyrealsense2 as rs
 
@@ -101,7 +101,7 @@ def save_images_from_cameras(
         serial_numbers = [cam["serial_number"] for cam in camera_infos]
 
     if mock:
-        import tests.mock_cv2 as cv2
+        import tests.cameras.mock_cv2 as cv2
     else:
         import cv2
 
@@ -252,7 +252,7 @@ class RealSenseCamera(Camera):
         self.logs = {}
 
         if self.mock:
-            import tests.mock_cv2 as cv2
+            import tests.cameras.mock_cv2 as cv2
         else:
             import cv2
 
@@ -284,7 +284,7 @@ class RealSenseCamera(Camera):
             raise DeviceAlreadyConnectedError(f"RealSenseCamera({self.serial_number}) is already connected.")
 
         if self.mock:
-            import tests.mock_pyrealsense2 as rs
+            import tests.cameras.mock_pyrealsense2 as rs
         else:
             import pyrealsense2 as rs
 
@@ -372,7 +372,7 @@ class RealSenseCamera(Camera):
             )
 
         if self.mock:
-            import tests.mock_cv2 as cv2
+            import tests.cameras.mock_cv2 as cv2
         else:
             import cv2
 
diff --git a/lerobot/common/cameras/opencv/camera_opencv.py b/lerobot/common/cameras/opencv/camera_opencv.py
index a7d5a0f3..f4bfb507 100644
--- a/lerobot/common/cameras/opencv/camera_opencv.py
+++ b/lerobot/common/cameras/opencv/camera_opencv.py
@@ -81,7 +81,7 @@ def _find_cameras(
     possible_camera_ids: list[int | str], raise_when_empty=False, mock=False
 ) -> list[int | str]:
     if mock:
-        import tests.mock_cv2 as cv2
+        import tests.cameras.mock_cv2 as cv2
     else:
         import cv2
 
@@ -270,7 +270,7 @@ class OpenCVCamera(Camera):
         self.logs = {}
 
         if self.mock:
-            import tests.mock_cv2 as cv2
+            import tests.cameras.mock_cv2 as cv2
         else:
             import cv2
 
@@ -287,7 +287,7 @@ class OpenCVCamera(Camera):
             raise DeviceAlreadyConnectedError(f"OpenCVCamera({self.camera_index}) is already connected.")
 
         if self.mock:
-            import tests.mock_cv2 as cv2
+            import tests.cameras.mock_cv2 as cv2
         else:
             import cv2
 
@@ -399,7 +399,7 @@ class OpenCVCamera(Camera):
         # so we convert the image color from BGR to RGB.
         if requested_color_mode == "rgb":
             if self.mock:
-                import tests.mock_cv2 as cv2
+                import tests.cameras.mock_cv2 as cv2
             else:
                 import cv2
 
diff --git a/lerobot/common/motors/dynamixel/dynamixel.py b/lerobot/common/motors/dynamixel/dynamixel.py
index 55f6069b..a09b1847 100644
--- a/lerobot/common/motors/dynamixel/dynamixel.py
+++ b/lerobot/common/motors/dynamixel/dynamixel.py
@@ -332,7 +332,7 @@ class DynamixelMotorsBus:
             )
 
         if self.mock:
-            import tests.mock_dynamixel_sdk as dxl
+            import tests.motors.mock_dynamixel_sdk as dxl
         else:
             import dynamixel_sdk as dxl
 
@@ -356,7 +356,7 @@ class DynamixelMotorsBus:
 
     def reconnect(self):
         if self.mock:
-            import tests.mock_dynamixel_sdk as dxl
+            import tests.motors.mock_dynamixel_sdk as dxl
         else:
             import dynamixel_sdk as dxl
 
@@ -646,7 +646,7 @@ class DynamixelMotorsBus:
 
     def read_with_motor_ids(self, motor_models, motor_ids, data_name, num_retry=NUM_READ_RETRY):
         if self.mock:
-            import tests.mock_dynamixel_sdk as dxl
+            import tests.motors.mock_dynamixel_sdk as dxl
         else:
             import dynamixel_sdk as dxl
 
@@ -691,7 +691,7 @@ class DynamixelMotorsBus:
         start_time = time.perf_counter()
 
         if self.mock:
-            import tests.mock_dynamixel_sdk as dxl
+            import tests.motors.mock_dynamixel_sdk as dxl
         else:
             import dynamixel_sdk as dxl
 
@@ -757,7 +757,7 @@ class DynamixelMotorsBus:
 
     def write_with_motor_ids(self, motor_models, motor_ids, data_name, values, num_retry=NUM_WRITE_RETRY):
         if self.mock:
-            import tests.mock_dynamixel_sdk as dxl
+            import tests.motors.mock_dynamixel_sdk as dxl
         else:
             import dynamixel_sdk as dxl
 
@@ -793,7 +793,7 @@ class DynamixelMotorsBus:
         start_time = time.perf_counter()
 
         if self.mock:
-            import tests.mock_dynamixel_sdk as dxl
+            import tests.motors.mock_dynamixel_sdk as dxl
         else:
             import dynamixel_sdk as dxl
 
diff --git a/lerobot/common/motors/feetech/feetech.py b/lerobot/common/motors/feetech/feetech.py
index c183967a..4a9d2cde 100644
--- a/lerobot/common/motors/feetech/feetech.py
+++ b/lerobot/common/motors/feetech/feetech.py
@@ -313,7 +313,7 @@ class FeetechMotorsBus:
             )
 
         if self.mock:
-            import tests.mock_scservo_sdk as scs
+            import tests.motors.mock_scservo_sdk as scs
         else:
             import scservo_sdk as scs
 
@@ -337,7 +337,7 @@ class FeetechMotorsBus:
 
     def reconnect(self):
         if self.mock:
-            import tests.mock_scservo_sdk as scs
+            import tests.motors.mock_scservo_sdk as scs
         else:
             import scservo_sdk as scs
 
@@ -664,7 +664,7 @@ class FeetechMotorsBus:
 
     def read_with_motor_ids(self, motor_models, motor_ids, data_name, num_retry=NUM_READ_RETRY):
         if self.mock:
-            import tests.mock_scservo_sdk as scs
+            import tests.motors.mock_scservo_sdk as scs
         else:
             import scservo_sdk as scs
 
@@ -702,7 +702,7 @@ class FeetechMotorsBus:
 
     def read(self, data_name, motor_names: str | list[str] | None = None):
         if self.mock:
-            import tests.mock_scservo_sdk as scs
+            import tests.motors.mock_scservo_sdk as scs
         else:
             import scservo_sdk as scs
 
@@ -782,7 +782,7 @@ class FeetechMotorsBus:
 
     def write_with_motor_ids(self, motor_models, motor_ids, data_name, values, num_retry=NUM_WRITE_RETRY):
         if self.mock:
-            import tests.mock_scservo_sdk as scs
+            import tests.motors.mock_scservo_sdk as scs
         else:
             import scservo_sdk as scs
 
@@ -818,7 +818,7 @@ class FeetechMotorsBus:
         start_time = time.perf_counter()
 
         if self.mock:
-            import tests.mock_scservo_sdk as scs
+            import tests.motors.mock_scservo_sdk as scs
         else:
             import scservo_sdk as scs
 
diff --git a/lerobot/common/robots/lekiwi/README.md b/lerobot/common/robots/lekiwi/README.md
index 4b811417..dc310af2 100644
--- a/lerobot/common/robots/lekiwi/README.md
+++ b/lerobot/common/robots/lekiwi/README.md
@@ -366,8 +366,8 @@ Now we have to calibrate the leader arm and the follower arm. The wheel motors d
 
 You will need to move the follower arm to these positions sequentially:
 
-| 1. Zero position | 2. Rotated position | 3. Rest position |
-|---|---|---|
+| 1. Zero position                                                                                                                                                  | 2. Rotated position                                                                                                                                                        | 3. Rest position                                                                                                                                                  |
+| ----------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | <img src="../media/lekiwi/mobile_calib_zero.webp?raw=true" alt="SO-100 follower arm zero position" title="SO-100 follower arm zero position" style="width:100%;"> | <img src="../media/lekiwi/mobile_calib_rotated.webp?raw=true" alt="SO-100 follower arm rotated position" title="SO-100 follower arm rotated position" style="width:100%;"> | <img src="../media/lekiwi/mobile_calib_rest.webp?raw=true" alt="SO-100 follower arm rest position" title="SO-100 follower arm rest position" style="width:100%;"> |
 
 Make sure the arm is connected to the Raspberry Pi and run this script (on the Raspberry Pi) to launch manual calibration:
@@ -385,8 +385,8 @@ If you have the **wired** LeKiwi version please run all commands including this
 ### Calibrate leader arm
 Then to calibrate the leader arm (which is attached to the laptop/pc). You will need to move the leader arm to these positions sequentially:
 
-| 1. Zero position | 2. Rotated position | 3. Rest position |
-|---|---|---|
+| 1. Zero position                                                                                                                                       | 2. Rotated position                                                                                                                                             | 3. Rest position                                                                                                                                       |
+| ------------------------------------------------------------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------ |
 | <img src="../media/so100/leader_zero.webp?raw=true" alt="SO-100 leader arm zero position" title="SO-100 leader arm zero position" style="width:100%;"> | <img src="../media/so100/leader_rotated.webp?raw=true" alt="SO-100 leader arm rotated position" title="SO-100 leader arm rotated position" style="width:100%;"> | <img src="../media/so100/leader_rest.webp?raw=true" alt="SO-100 leader arm rest position" title="SO-100 leader arm rest position" style="width:100%;"> |
 
 Run this script (on your laptop/pc) to launch manual calibration:
@@ -416,22 +416,22 @@ python lerobot/scripts/control_robot.py \
 
 You should see on your laptop something like this: ```[INFO] Connected to remote robot at tcp://172.17.133.91:5555 and video stream at tcp://172.17.133.91:5556.``` Now you can move the leader arm and use the keyboard (w,a,s,d) to drive forward, left, backwards, right. And use (z,x) to turn left or turn right. You can use (r,f) to increase and decrease the speed of the mobile robot. There are three speed modes, see the table below:
 | Speed Mode | Linear Speed (m/s) | Rotation Speed (deg/s) |
-|------------|-------------------|-----------------------|
-| Fast      | 0.4               | 90                    |
-| Medium    | 0.25              | 60                    |
-| Slow      | 0.1               | 30                    |
+| ---------- | ------------------ | ---------------------- |
+| Fast       | 0.4                | 90                     |
+| Medium     | 0.25               | 60                     |
+| Slow       | 0.1                | 30                     |
 
 
-| Key  | Action                         |
-|------|--------------------------------|
-| W    | Move forward                   |
-| A    | Move left                       |
-| S    | Move backward                   |
-| D    | Move right                      |
-| Z    | Turn left                       |
-| X    | Turn right                      |
-| R    | Increase speed                  |
-| F    | Decrease speed                  |
+| Key | Action         |
+| --- | -------------- |
+| W   | Move forward   |
+| A   | Move left      |
+| S   | Move backward  |
+| D   | Move right     |
+| Z   | Turn left      |
+| X   | Turn right     |
+| R   | Increase speed |
+| F   | Decrease speed |
 
 > [!TIP]
 >  If you use a different keyboard you can change the keys for each command in the [`LeKiwiRobotConfig`](../lerobot/common/robot_devices/robots/configs.py).
@@ -549,14 +549,14 @@ python lerobot/scripts/train.py \
   --policy.type=act \
   --output_dir=outputs/train/act_lekiwi_test \
   --job_name=act_lekiwi_test \
-  --device=cuda \
+  --policy.device=cuda \
   --wandb.enable=true
 ```
 
 Let's explain it:
 1. We provided the dataset as argument with `--dataset.repo_id=${HF_USER}/lekiwi_test`.
 2. We provided the policy with `policy.type=act`. This loads configurations from [`configuration_act.py`](../lerobot/common/policies/act/configuration_act.py). Importantly, this policy will automatically adapt to the number of motor sates, motor actions and cameras of your robot (e.g. `laptop` and `phone`) which have been saved in your dataset.
-4. We provided `device=cuda` since we are training on a Nvidia GPU, but you could use `device=mps` to train on Apple silicon.
+4. We provided `policy.device=cuda` since we are training on a Nvidia GPU, but you could use `policy.device=mps` to train on Apple silicon.
 5. We provided `wandb.enable=true` to use [Weights and Biases](https://docs.wandb.ai/quickstart) for visualizing training plots. This is optional but if you use it, make sure you are logged in by running `wandb login`.
 
 Training should take several hours. You will find checkpoints in `outputs/train/act_lekiwi_test/checkpoints`.
diff --git a/lerobot/common/robots/moss/README.md b/lerobot/common/robots/moss/README.md
index 67f8157e..d2e02076 100644
--- a/lerobot/common/robots/moss/README.md
+++ b/lerobot/common/robots/moss/README.md
@@ -176,8 +176,8 @@ Next, you'll need to calibrate your Moss v1 robot to ensure that the leader and
 
 You will need to move the follower arm to these positions sequentially:
 
-| 1. Zero position | 2. Rotated position | 3. Rest position |
-|---|---|---|
+| 1. Zero position                                                                                                                                              | 2. Rotated position                                                                                                                                                    | 3. Rest position                                                                                                                                              |
+| ------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | <img src="../media/moss/follower_zero.webp?raw=true" alt="Moss v1 follower arm zero position" title="Moss v1 follower arm zero position" style="width:100%;"> | <img src="../media/moss/follower_rotated.webp?raw=true" alt="Moss v1 follower arm rotated position" title="Moss v1 follower arm rotated position" style="width:100%;"> | <img src="../media/moss/follower_rest.webp?raw=true" alt="Moss v1 follower arm rest position" title="Moss v1 follower arm rest position" style="width:100%;"> |
 
 Make sure both arms are connected and run this script to launch manual calibration:
@@ -192,8 +192,8 @@ python lerobot/scripts/control_robot.py \
 **Manual calibration of leader arm**
 Follow step 6 of the [assembly video](https://www.youtube.com/watch?v=DA91NJOtMic) which illustrates the manual calibration. You will need to move the leader arm to these positions sequentially:
 
-| 1. Zero position | 2. Rotated position | 3. Rest position |
-|---|---|---|
+| 1. Zero position                                                                                                                                        | 2. Rotated position                                                                                                                                              | 3. Rest position                                                                                                                                        |
+| ------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | <img src="../media/moss/leader_zero.webp?raw=true" alt="Moss v1 leader arm zero position" title="Moss v1 leader arm zero position" style="width:100%;"> | <img src="../media/moss/leader_rotated.webp?raw=true" alt="Moss v1 leader arm rotated position" title="Moss v1 leader arm rotated position" style="width:100%;"> | <img src="../media/moss/leader_rest.webp?raw=true" alt="Moss v1 leader arm rest position" title="Moss v1 leader arm rest position" style="width:100%;"> |
 
 Run this script to launch manual calibration:
@@ -293,14 +293,14 @@ python lerobot/scripts/train.py \
   --policy.type=act \
   --output_dir=outputs/train/act_moss_test \
   --job_name=act_moss_test \
-  --device=cuda \
+  --policy.device=cuda \
   --wandb.enable=true
 ```
 
 Let's explain it:
 1. We provided the dataset as argument with `--dataset.repo_id=${HF_USER}/moss_test`.
 2. We provided the policy with `policy.type=act`. This loads configurations from [`configuration_act.py`](../lerobot/common/policies/act/configuration_act.py). Importantly, this policy will automatically adapt to the number of motor sates, motor actions and cameras of your robot (e.g. `laptop` and `phone`) which have been saved in your dataset.
-4. We provided `device=cuda` since we are training on a Nvidia GPU, but you could use `device=mps` to train on Apple silicon.
+4. We provided `policy.device=cuda` since we are training on a Nvidia GPU, but you could use `policy.device=mps` to train on Apple silicon.
 5. We provided `wandb.enable=true` to use [Weights and Biases](https://docs.wandb.ai/quickstart) for visualizing training plots. This is optional but if you use it, make sure you are logged in by running `wandb login`.
 
 Training should take several hours. You will find checkpoints in `outputs/train/act_moss_test/checkpoints`.
diff --git a/lerobot/common/robots/so100/README.md b/lerobot/common/robots/so100/README.md
index b63eb146..b8b45aa5 100644
--- a/lerobot/common/robots/so100/README.md
+++ b/lerobot/common/robots/so100/README.md
@@ -454,8 +454,8 @@ Next, you'll need to calibrate your SO-100 robot to ensure that the leader and f
 
 You will need to move the follower arm to these positions sequentially:
 
-| 1. Zero position | 2. Rotated position | 3. Rest position |
-|---|---|---|
+| 1. Zero position                                                                                                                                             | 2. Rotated position                                                                                                                                                   | 3. Rest position                                                                                                                                             |
+| ------------------------------------------------------------------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------ |
 | <img src="../media/so100/follower_zero.webp?raw=true" alt="SO-100 follower arm zero position" title="SO-100 follower arm zero position" style="width:100%;"> | <img src="../media/so100/follower_rotated.webp?raw=true" alt="SO-100 follower arm rotated position" title="SO-100 follower arm rotated position" style="width:100%;"> | <img src="../media/so100/follower_rest.webp?raw=true" alt="SO-100 follower arm rest position" title="SO-100 follower arm rest position" style="width:100%;"> |
 
 Make sure both arms are connected and run this script to launch manual calibration:
@@ -470,8 +470,8 @@ python lerobot/scripts/control_robot.py \
 #### b. Manual calibration of leader arm
 Follow step 6 of the [assembly video](https://youtu.be/FioA2oeFZ5I?t=724) which illustrates the manual calibration. You will need to move the leader arm to these positions sequentially:
 
-| 1. Zero position | 2. Rotated position | 3. Rest position |
-|---|---|---|
+| 1. Zero position                                                                                                                                       | 2. Rotated position                                                                                                                                             | 3. Rest position                                                                                                                                       |
+| ------------------------------------------------------------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------ |
 | <img src="../media/so100/leader_zero.webp?raw=true" alt="SO-100 leader arm zero position" title="SO-100 leader arm zero position" style="width:100%;"> | <img src="../media/so100/leader_rotated.webp?raw=true" alt="SO-100 leader arm rotated position" title="SO-100 leader arm rotated position" style="width:100%;"> | <img src="../media/so100/leader_rest.webp?raw=true" alt="SO-100 leader arm rest position" title="SO-100 leader arm rest position" style="width:100%;"> |
 
 Run this script to launch manual calibration:
@@ -571,14 +571,14 @@ python lerobot/scripts/train.py \
   --policy.type=act \
   --output_dir=outputs/train/act_so100_test \
   --job_name=act_so100_test \
-  --device=cuda \
+  --policy.device=cuda \
   --wandb.enable=true
 ```
 
 Let's explain it:
 1. We provided the dataset as argument with `--dataset.repo_id=${HF_USER}/so100_test`.
 2. We provided the policy with `policy.type=act`. This loads configurations from [`configuration_act.py`](../lerobot/common/policies/act/configuration_act.py). Importantly, this policy will automatically adapt to the number of motor sates, motor actions and cameras of your robot (e.g. `laptop` and `phone`) which have been saved in your dataset.
-4. We provided `device=cuda` since we are training on a Nvidia GPU, but you could use `device=mps` to train on Apple silicon.
+4. We provided `policy.device=cuda` since we are training on a Nvidia GPU, but you could use `policy.device=mps` to train on Apple silicon.
 5. We provided `wandb.enable=true` to use [Weights and Biases](https://docs.wandb.ai/quickstart) for visualizing training plots. This is optional but if you use it, make sure you are logged in by running `wandb login`.
 
 Training should take several hours. You will find checkpoints in `outputs/train/act_so100_test/checkpoints`.
diff --git a/lerobot/common/robots/viperx/README.md b/lerobot/common/robots/viperx/README.md
index 055551f0..62dee588 100644
--- a/lerobot/common/robots/viperx/README.md
+++ b/lerobot/common/robots/viperx/README.md
@@ -135,14 +135,14 @@ python lerobot/scripts/train.py \
   --policy.type=act \
   --output_dir=outputs/train/act_aloha_test \
   --job_name=act_aloha_test \
-  --device=cuda \
+  --policy.device=cuda \
   --wandb.enable=true
 ```
 
 Let's explain it:
 1. We provided the dataset as argument with `--dataset.repo_id=${HF_USER}/aloha_test`.
 2. We provided the policy with `policy.type=act`. This loads configurations from [`configuration_act.py`](../lerobot/common/policies/act/configuration_act.py). Importantly, this policy will automatically adapt to the number of motor sates, motor actions and cameras of your robot (e.g. `laptop` and `phone`) which have been saved in your dataset.
-4. We provided `device=cuda` since we are training on a Nvidia GPU, but you could use `device=mps` to train on Apple silicon.
+4. We provided `policy.device=cuda` since we are training on a Nvidia GPU, but you could use `policy.device=mps` to train on Apple silicon.
 5. We provided `wandb.enable=true` to use [Weights and Biases](https://docs.wandb.ai/quickstart) for visualizing training plots. This is optional but if you use it, make sure you are logged in by running `wandb login`.
 
 For more information on the `train` script see the previous tutorial: [`examples/4_train_policy_with_script.md`](../examples/4_train_policy_with_script.md)
diff --git a/lerobot/scripts/visualize_dataset.py b/lerobot/scripts/visualize_dataset.py
index 11feb1af..cdfea6b8 100644
--- a/lerobot/scripts/visualize_dataset.py
+++ b/lerobot/scripts/visualize_dataset.py
@@ -265,13 +265,25 @@ def main():
         ),
     )
 
+    parser.add_argument(
+        "--tolerance-s",
+        type=float,
+        default=1e-4,
+        help=(
+            "Tolerance in seconds used to ensure data timestamps respect the dataset fps value"
+            "This is argument passed to the constructor of LeRobotDataset and maps to its tolerance_s constructor argument"
+            "If not given, defaults to 1e-4."
+        ),
+    )
+
     args = parser.parse_args()
     kwargs = vars(args)
     repo_id = kwargs.pop("repo_id")
     root = kwargs.pop("root")
+    tolerance_s = kwargs.pop("tolerance_s")
 
     logging.info("Loading dataset")
-    dataset = LeRobotDataset(repo_id, root=root)
+    dataset = LeRobotDataset(repo_id, root=root, tolerance_s=tolerance_s)
 
     visualize_dataset(dataset, **vars(args))
 
diff --git a/lerobot/scripts/visualize_dataset_html.py b/lerobot/scripts/visualize_dataset_html.py
index f944144a..0fc21a8f 100644
--- a/lerobot/scripts/visualize_dataset_html.py
+++ b/lerobot/scripts/visualize_dataset_html.py
@@ -446,15 +446,31 @@ def main():
         help="Delete the output directory if it exists already.",
     )
 
+    parser.add_argument(
+        "--tolerance-s",
+        type=float,
+        default=1e-4,
+        help=(
+            "Tolerance in seconds used to ensure data timestamps respect the dataset fps value"
+            "This is argument passed to the constructor of LeRobotDataset and maps to its tolerance_s constructor argument"
+            "If not given, defaults to 1e-4."
+        ),
+    )
+
     args = parser.parse_args()
     kwargs = vars(args)
     repo_id = kwargs.pop("repo_id")
     load_from_hf_hub = kwargs.pop("load_from_hf_hub")
     root = kwargs.pop("root")
+    tolerance_s = kwargs.pop("tolerance_s")
 
     dataset = None
     if repo_id:
-        dataset = LeRobotDataset(repo_id, root=root) if not load_from_hf_hub else get_dataset_info(repo_id)
+        dataset = (
+            LeRobotDataset(repo_id, root=root, tolerance_s=tolerance_s)
+            if not load_from_hf_hub
+            else get_dataset_info(repo_id)
+        )
 
     visualize_dataset_html(dataset, **vars(args))
 
diff --git a/pyproject.toml b/pyproject.toml
index a4458eb6..19a5cffa 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -56,7 +56,6 @@ dependencies = [
     "gymnasium==0.29.1",                                                 # TODO(rcadene, aliberts): Make gym 1.0.0 work
     "h5py>=3.10.0",
     "huggingface-hub[hf-transfer,cli]>=0.27.1 ; python_version < '4.0'",
-    "hydra-core>=1.3.2",
     "imageio[ffmpeg]>=2.34.0",
     "jsonlines>=4.0.0",
     "numba>=0.59.0",
@@ -103,30 +102,7 @@ requires-poetry = ">=2.1"
 [tool.ruff]
 line-length = 110
 target-version = "py310"
-exclude = [
-    "tests/data",
-    ".bzr",
-    ".direnv",
-    ".eggs",
-    ".git",
-    ".git-rewrite",
-    ".hg",
-    ".mypy_cache",
-    ".nox",
-    ".pants.d",
-    ".pytype",
-    ".ruff_cache",
-    ".svn",
-    ".tox",
-    ".venv",
-    "__pypackages__",
-    "_build",
-    "buck-out",
-    "build",
-    "dist",
-    "node_modules",
-    "venv",
-]
+exclude = ["tests/artifacts/**/*.safetensors"]
 
 [tool.ruff.lint]
 select = ["E4", "E7", "E9", "F", "I", "N", "B", "C4", "SIM"]
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_0.safetensors b/tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_0.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_0.safetensors
rename to tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_0.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_1.safetensors b/tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_1.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_1.safetensors
rename to tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_1.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_250.safetensors b/tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_250.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_250.safetensors
rename to tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_250.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_251.safetensors b/tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_251.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_251.safetensors
rename to tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_251.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_498.safetensors b/tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_498.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_498.safetensors
rename to tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_498.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_499.safetensors b/tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_499.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/aloha_sim_insertion_human/frame_499.safetensors
rename to tests/artifacts/datasets/lerobot/aloha_sim_insertion_human/frame_499.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_0.safetensors b/tests/artifacts/datasets/lerobot/pusht/frame_0.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_0.safetensors
rename to tests/artifacts/datasets/lerobot/pusht/frame_0.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_1.safetensors b/tests/artifacts/datasets/lerobot/pusht/frame_1.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_1.safetensors
rename to tests/artifacts/datasets/lerobot/pusht/frame_1.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_159.safetensors b/tests/artifacts/datasets/lerobot/pusht/frame_159.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_159.safetensors
rename to tests/artifacts/datasets/lerobot/pusht/frame_159.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_160.safetensors b/tests/artifacts/datasets/lerobot/pusht/frame_160.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_160.safetensors
rename to tests/artifacts/datasets/lerobot/pusht/frame_160.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_80.safetensors b/tests/artifacts/datasets/lerobot/pusht/frame_80.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_80.safetensors
rename to tests/artifacts/datasets/lerobot/pusht/frame_80.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_81.safetensors b/tests/artifacts/datasets/lerobot/pusht/frame_81.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/pusht/frame_81.safetensors
rename to tests/artifacts/datasets/lerobot/pusht/frame_81.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_0.safetensors b/tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_0.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_0.safetensors
rename to tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_0.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_1.safetensors b/tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_1.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_1.safetensors
rename to tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_1.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_12.safetensors b/tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_12.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_12.safetensors
rename to tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_12.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_13.safetensors b/tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_13.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_13.safetensors
rename to tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_13.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_23.safetensors b/tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_23.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_23.safetensors
rename to tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_23.safetensors
diff --git a/tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_24.safetensors b/tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_24.safetensors
similarity index 100%
rename from tests/data/save_dataset_to_safetensors/lerobot/xarm_lift_medium/frame_24.safetensors
rename to tests/artifacts/datasets/lerobot/xarm_lift_medium/frame_24.safetensors
diff --git a/tests/scripts/save_dataset_to_safetensors.py b/tests/artifacts/datasets/save_dataset_to_safetensors.py
similarity index 95%
rename from tests/scripts/save_dataset_to_safetensors.py
rename to tests/artifacts/datasets/save_dataset_to_safetensors.py
index 3b77348c..74d42a3d 100644
--- a/tests/scripts/save_dataset_to_safetensors.py
+++ b/tests/artifacts/datasets/save_dataset_to_safetensors.py
@@ -23,7 +23,7 @@ If you know that your change will break backward compatibility, you should write
 doesnt need to be merged into the `main` branch. Then you need to run this script and update the tests artifacts.
 
 Example usage:
-    `python tests/scripts/save_dataset_to_safetensors.py`
+    `python tests/artifacts/datasets/save_dataset_to_safetensors.py`
 """
 
 import shutil
@@ -88,4 +88,4 @@ if __name__ == "__main__":
         "lerobot/nyu_franka_play_dataset",
         "lerobot/cmu_stretch",
     ]:
-        save_dataset_to_safetensors("tests/data/save_dataset_to_safetensors", repo_id=dataset)
+        save_dataset_to_safetensors("tests/artifacts/datasets", repo_id=dataset)
diff --git a/tests/data/save_image_transforms_to_safetensors/default_transforms.safetensors b/tests/artifacts/image_transforms/default_transforms.safetensors
similarity index 100%
rename from tests/data/save_image_transforms_to_safetensors/default_transforms.safetensors
rename to tests/artifacts/image_transforms/default_transforms.safetensors
diff --git a/tests/scripts/save_image_transforms_to_safetensors.py b/tests/artifacts/image_transforms/save_image_transforms_to_safetensors.py
similarity index 97%
rename from tests/scripts/save_image_transforms_to_safetensors.py
rename to tests/artifacts/image_transforms/save_image_transforms_to_safetensors.py
index bd2c3add..7b037af4 100644
--- a/tests/scripts/save_image_transforms_to_safetensors.py
+++ b/tests/artifacts/image_transforms/save_image_transforms_to_safetensors.py
@@ -27,7 +27,7 @@ from lerobot.common.datasets.transforms import (
 )
 from lerobot.common.utils.random_utils import seeded_context
 
-ARTIFACT_DIR = Path("tests/data/save_image_transforms_to_safetensors")
+ARTIFACT_DIR = Path("tests/artifacts/image_transforms")
 DATASET_REPO_ID = "lerobot/aloha_mobile_shrimp"
 
 
diff --git a/tests/data/save_image_transforms_to_safetensors/single_transforms.safetensors b/tests/artifacts/image_transforms/single_transforms.safetensors
similarity index 100%
rename from tests/data/save_image_transforms_to_safetensors/single_transforms.safetensors
rename to tests/artifacts/image_transforms/single_transforms.safetensors
diff --git a/tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_/actions.safetensors b/tests/artifacts/policies/aloha_sim_insertion_human_act_/actions.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_/actions.safetensors
rename to tests/artifacts/policies/aloha_sim_insertion_human_act_/actions.safetensors
diff --git a/tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_/grad_stats.safetensors b/tests/artifacts/policies/aloha_sim_insertion_human_act_/grad_stats.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_/grad_stats.safetensors
rename to tests/artifacts/policies/aloha_sim_insertion_human_act_/grad_stats.safetensors
diff --git a/tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_/output_dict.safetensors b/tests/artifacts/policies/aloha_sim_insertion_human_act_/output_dict.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_/output_dict.safetensors
rename to tests/artifacts/policies/aloha_sim_insertion_human_act_/output_dict.safetensors
diff --git a/tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_/param_stats.safetensors b/tests/artifacts/policies/aloha_sim_insertion_human_act_/param_stats.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_/param_stats.safetensors
rename to tests/artifacts/policies/aloha_sim_insertion_human_act_/param_stats.safetensors
diff --git a/tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_1000_steps/actions.safetensors b/tests/artifacts/policies/aloha_sim_insertion_human_act_1000_steps/actions.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_1000_steps/actions.safetensors
rename to tests/artifacts/policies/aloha_sim_insertion_human_act_1000_steps/actions.safetensors
diff --git a/tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_1000_steps/grad_stats.safetensors b/tests/artifacts/policies/aloha_sim_insertion_human_act_1000_steps/grad_stats.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_1000_steps/grad_stats.safetensors
rename to tests/artifacts/policies/aloha_sim_insertion_human_act_1000_steps/grad_stats.safetensors
diff --git a/tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_1000_steps/output_dict.safetensors b/tests/artifacts/policies/aloha_sim_insertion_human_act_1000_steps/output_dict.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_1000_steps/output_dict.safetensors
rename to tests/artifacts/policies/aloha_sim_insertion_human_act_1000_steps/output_dict.safetensors
diff --git a/tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_1000_steps/param_stats.safetensors b/tests/artifacts/policies/aloha_sim_insertion_human_act_1000_steps/param_stats.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/aloha_sim_insertion_human_act_1000_steps/param_stats.safetensors
rename to tests/artifacts/policies/aloha_sim_insertion_human_act_1000_steps/param_stats.safetensors
diff --git a/tests/data/save_policy_to_safetensors/pusht_diffusion_/actions.safetensors b/tests/artifacts/policies/pusht_diffusion_/actions.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/pusht_diffusion_/actions.safetensors
rename to tests/artifacts/policies/pusht_diffusion_/actions.safetensors
diff --git a/tests/data/save_policy_to_safetensors/pusht_diffusion_/grad_stats.safetensors b/tests/artifacts/policies/pusht_diffusion_/grad_stats.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/pusht_diffusion_/grad_stats.safetensors
rename to tests/artifacts/policies/pusht_diffusion_/grad_stats.safetensors
diff --git a/tests/data/save_policy_to_safetensors/pusht_diffusion_/output_dict.safetensors b/tests/artifacts/policies/pusht_diffusion_/output_dict.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/pusht_diffusion_/output_dict.safetensors
rename to tests/artifacts/policies/pusht_diffusion_/output_dict.safetensors
diff --git a/tests/data/save_policy_to_safetensors/pusht_diffusion_/param_stats.safetensors b/tests/artifacts/policies/pusht_diffusion_/param_stats.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/pusht_diffusion_/param_stats.safetensors
rename to tests/artifacts/policies/pusht_diffusion_/param_stats.safetensors
diff --git a/tests/scripts/save_policy_to_safetensors.py b/tests/artifacts/policies/save_policy_to_safetensors.py
similarity index 98%
rename from tests/scripts/save_policy_to_safetensors.py
rename to tests/artifacts/policies/save_policy_to_safetensors.py
index 60fd9fc0..106f0dc0 100644
--- a/tests/scripts/save_policy_to_safetensors.py
+++ b/tests/artifacts/policies/save_policy_to_safetensors.py
@@ -141,5 +141,5 @@ if __name__ == "__main__":
         raise RuntimeError("No policies were provided!")
     for ds_repo_id, policy, policy_kwargs, file_name_extra in artifacts_cfg:
         ds_name = ds_repo_id.split("/")[-1]
-        output_dir = Path("tests/data/save_policy_to_safetensors") / f"{ds_name}_{policy}_{file_name_extra}"
+        output_dir = Path("tests/artifacts/policies") / f"{ds_name}_{policy}_{file_name_extra}"
         save_policy_to_safetensors(output_dir, ds_repo_id, policy, policy_kwargs)
diff --git a/tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_mpc/actions.safetensors b/tests/artifacts/policies/xarm_lift_medium_tdmpc_use_mpc/actions.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_mpc/actions.safetensors
rename to tests/artifacts/policies/xarm_lift_medium_tdmpc_use_mpc/actions.safetensors
diff --git a/tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_mpc/grad_stats.safetensors b/tests/artifacts/policies/xarm_lift_medium_tdmpc_use_mpc/grad_stats.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_mpc/grad_stats.safetensors
rename to tests/artifacts/policies/xarm_lift_medium_tdmpc_use_mpc/grad_stats.safetensors
diff --git a/tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_mpc/output_dict.safetensors b/tests/artifacts/policies/xarm_lift_medium_tdmpc_use_mpc/output_dict.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_mpc/output_dict.safetensors
rename to tests/artifacts/policies/xarm_lift_medium_tdmpc_use_mpc/output_dict.safetensors
diff --git a/tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_mpc/param_stats.safetensors b/tests/artifacts/policies/xarm_lift_medium_tdmpc_use_mpc/param_stats.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_mpc/param_stats.safetensors
rename to tests/artifacts/policies/xarm_lift_medium_tdmpc_use_mpc/param_stats.safetensors
diff --git a/tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_policy/actions.safetensors b/tests/artifacts/policies/xarm_lift_medium_tdmpc_use_policy/actions.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_policy/actions.safetensors
rename to tests/artifacts/policies/xarm_lift_medium_tdmpc_use_policy/actions.safetensors
diff --git a/tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_policy/grad_stats.safetensors b/tests/artifacts/policies/xarm_lift_medium_tdmpc_use_policy/grad_stats.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_policy/grad_stats.safetensors
rename to tests/artifacts/policies/xarm_lift_medium_tdmpc_use_policy/grad_stats.safetensors
diff --git a/tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_policy/output_dict.safetensors b/tests/artifacts/policies/xarm_lift_medium_tdmpc_use_policy/output_dict.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_policy/output_dict.safetensors
rename to tests/artifacts/policies/xarm_lift_medium_tdmpc_use_policy/output_dict.safetensors
diff --git a/tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_policy/param_stats.safetensors b/tests/artifacts/policies/xarm_lift_medium_tdmpc_use_policy/param_stats.safetensors
similarity index 100%
rename from tests/data/save_policy_to_safetensors/xarm_lift_medium_tdmpc_use_policy/param_stats.safetensors
rename to tests/artifacts/policies/xarm_lift_medium_tdmpc_use_policy/param_stats.safetensors
diff --git a/tests/mock_cv2.py b/tests/cameras/mock_cv2.py
similarity index 100%
rename from tests/mock_cv2.py
rename to tests/cameras/mock_cv2.py
diff --git a/tests/mock_pyrealsense2.py b/tests/cameras/mock_pyrealsense2.py
similarity index 100%
rename from tests/mock_pyrealsense2.py
rename to tests/cameras/mock_pyrealsense2.py
diff --git a/tests/test_cameras.py b/tests/cameras/test_cameras.py
similarity index 99%
rename from tests/test_cameras.py
rename to tests/cameras/test_cameras.py
index 9cce74c4..6dbc716c 100644
--- a/tests/test_cameras.py
+++ b/tests/cameras/test_cameras.py
@@ -146,7 +146,7 @@ def test_camera(request, camera_type, mock):
         camera.connect()
 
         if mock:
-            import tests.mock_cv2 as cv2
+            import tests.cameras.mock_cv2 as cv2
         else:
             import cv2
 
diff --git a/tests/test_compute_stats.py b/tests/datasets/test_compute_stats.py
similarity index 100%
rename from tests/test_compute_stats.py
rename to tests/datasets/test_compute_stats.py
diff --git a/tests/test_datasets.py b/tests/datasets/test_datasets.py
similarity index 99%
rename from tests/test_datasets.py
rename to tests/datasets/test_datasets.py
index 1a321e24..2b0fc5b1 100644
--- a/tests/test_datasets.py
+++ b/tests/datasets/test_datasets.py
@@ -473,12 +473,12 @@ def test_flatten_unflatten_dict():
 )
 @require_x86_64_kernel
 def test_backward_compatibility(repo_id):
-    """The artifacts for this test have been generated by `tests/scripts/save_dataset_to_safetensors.py`."""
+    """The artifacts for this test have been generated by `tests/artifacts/datasets/save_dataset_to_safetensors.py`."""
 
     # TODO(rcadene, aliberts): remove dataset download
     dataset = LeRobotDataset(repo_id, episodes=[0])
 
-    test_dir = Path("tests/data/save_dataset_to_safetensors") / repo_id
+    test_dir = Path("tests/artifacts/datasets") / repo_id
 
     def load_and_compare(i):
         new_frame = dataset[i]  # noqa: B023
diff --git a/tests/test_delta_timestamps.py b/tests/datasets/test_delta_timestamps.py
similarity index 100%
rename from tests/test_delta_timestamps.py
rename to tests/datasets/test_delta_timestamps.py
diff --git a/tests/test_image_transforms.py b/tests/datasets/test_image_transforms.py
similarity index 99%
rename from tests/test_image_transforms.py
rename to tests/datasets/test_image_transforms.py
index 19bd77df..352aba99 100644
--- a/tests/test_image_transforms.py
+++ b/tests/datasets/test_image_transforms.py
@@ -33,7 +33,7 @@ from lerobot.scripts.visualize_image_transforms import (
     save_all_transforms,
     save_each_transform,
 )
-from tests.scripts.save_image_transforms_to_safetensors import ARTIFACT_DIR
+from tests.artifacts.image_transforms.save_image_transforms_to_safetensors import ARTIFACT_DIR
 from tests.utils import require_x86_64_kernel
 
 
diff --git a/tests/test_image_writer.py b/tests/datasets/test_image_writer.py
similarity index 100%
rename from tests/test_image_writer.py
rename to tests/datasets/test_image_writer.py
diff --git a/tests/test_online_buffer.py b/tests/datasets/test_online_buffer.py
similarity index 100%
rename from tests/test_online_buffer.py
rename to tests/datasets/test_online_buffer.py
diff --git a/tests/test_sampler.py b/tests/datasets/test_sampler.py
similarity index 100%
rename from tests/test_sampler.py
rename to tests/datasets/test_sampler.py
diff --git a/tests/test_utils.py b/tests/datasets/test_utils.py
similarity index 65%
rename from tests/test_utils.py
rename to tests/datasets/test_utils.py
index 2d0efc5a..0d02218a 100644
--- a/tests/test_utils.py
+++ b/tests/datasets/test_utils.py
@@ -1,3 +1,5 @@
+#!/usr/bin/env python
+
 # Copyright 2024 The HuggingFace Inc. team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
@@ -11,13 +13,32 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+
 import torch
 from datasets import Dataset
+from huggingface_hub import DatasetCard
 
 from lerobot.common.datasets.push_dataset_to_hub.utils import calculate_episode_data_index
-from lerobot.common.datasets.utils import (
-    hf_transform_to_torch,
-)
+from lerobot.common.datasets.utils import create_lerobot_dataset_card, hf_transform_to_torch
+
+
+def test_default_parameters():
+    card = create_lerobot_dataset_card()
+    assert isinstance(card, DatasetCard)
+    assert card.data.tags == ["LeRobot"]
+    assert card.data.task_categories == ["robotics"]
+    assert card.data.configs == [
+        {
+            "config_name": "default",
+            "data_files": "data/*/*.parquet",
+        }
+    ]
+
+
+def test_with_tags():
+    tags = ["tag1", "tag2"]
+    card = create_lerobot_dataset_card(tags=tags)
+    assert card.data.tags == ["LeRobot", "tag1", "tag2"]
 
 
 def test_calculate_episode_data_index():
diff --git a/tests/test_visualize_dataset.py b/tests/datasets/test_visualize_dataset.py
similarity index 100%
rename from tests/test_visualize_dataset.py
rename to tests/datasets/test_visualize_dataset.py
diff --git a/tests/test_envs.py b/tests/envs/test_envs.py
similarity index 98%
rename from tests/test_envs.py
rename to tests/envs/test_envs.py
index c7c384db..b318abb4 100644
--- a/tests/test_envs.py
+++ b/tests/envs/test_envs.py
@@ -23,8 +23,7 @@ from gymnasium.utils.env_checker import check_env
 import lerobot
 from lerobot.common.envs.factory import make_env, make_env_config
 from lerobot.common.envs.utils import preprocess_observation
-
-from .utils import require_env
+from tests.utils import require_env
 
 OBS_TYPES = ["state", "pixels", "pixels_agent_pos"]
 
diff --git a/tests/test_examples.py b/tests/examples/test_examples.py
similarity index 100%
rename from tests/test_examples.py
rename to tests/examples/test_examples.py
diff --git a/tests/lerobot/common/datasets/test_utils.py b/tests/lerobot/common/datasets/test_utils.py
deleted file mode 100644
index f484e1ae..00000000
--- a/tests/lerobot/common/datasets/test_utils.py
+++ /dev/null
@@ -1,38 +0,0 @@
-#!/usr/bin/env python
-
-# Copyright 2024 The HuggingFace Inc. team. All rights reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-
-from huggingface_hub import DatasetCard
-
-from lerobot.common.datasets.utils import create_lerobot_dataset_card
-
-
-def test_default_parameters():
-    card = create_lerobot_dataset_card()
-    assert isinstance(card, DatasetCard)
-    assert card.data.tags == ["LeRobot"]
-    assert card.data.task_categories == ["robotics"]
-    assert card.data.configs == [
-        {
-            "config_name": "default",
-            "data_files": "data/*/*.parquet",
-        }
-    ]
-
-
-def test_with_tags():
-    tags = ["tag1", "tag2"]
-    card = create_lerobot_dataset_card(tags=tags)
-    assert card.data.tags == ["LeRobot", "tag1", "tag2"]
diff --git a/tests/mock_dynamixel_sdk.py b/tests/motors/mock_dynamixel_sdk.py
similarity index 100%
rename from tests/mock_dynamixel_sdk.py
rename to tests/motors/mock_dynamixel_sdk.py
diff --git a/tests/mock_scservo_sdk.py b/tests/motors/mock_scservo_sdk.py
similarity index 100%
rename from tests/mock_scservo_sdk.py
rename to tests/motors/mock_scservo_sdk.py
diff --git a/tests/test_motors.py b/tests/motors/test_motors.py
similarity index 100%
rename from tests/test_motors.py
rename to tests/motors/test_motors.py
diff --git a/tests/test_optimizers.py b/tests/optim/test_optimizers.py
similarity index 100%
rename from tests/test_optimizers.py
rename to tests/optim/test_optimizers.py
diff --git a/tests/test_schedulers.py b/tests/optim/test_schedulers.py
similarity index 100%
rename from tests/test_schedulers.py
rename to tests/optim/test_schedulers.py
diff --git a/tests/test_policies.py b/tests/policies/test_policies.py
similarity index 98%
rename from tests/test_policies.py
rename to tests/policies/test_policies.py
index 5d7cca8f..197aa732 100644
--- a/tests/test_policies.py
+++ b/tests/policies/test_policies.py
@@ -40,7 +40,7 @@ from lerobot.common.utils.random_utils import seeded_context
 from lerobot.configs.default import DatasetConfig
 from lerobot.configs.train import TrainPipelineConfig
 from lerobot.configs.types import FeatureType, NormalizationMode, PolicyFeature
-from tests.scripts.save_policy_to_safetensors import get_policy_stats
+from tests.artifacts.policies.save_policy_to_safetensors import get_policy_stats
 from tests.utils import DEVICE, require_cpu, require_env, require_x86_64_kernel
 
 
@@ -407,12 +407,10 @@ def test_backward_compatibility(ds_repo_id: str, policy_name: str, policy_kwargs
            should be updated.
         4. Check that this test now passes.
         5. Remember to restore `tests/scripts/save_policy_to_safetensors.py` to its original state.
-        6. Remember to stage and commit the resulting changes to `tests/data`.
+        6. Remember to stage and commit the resulting changes to `tests/artifacts`.
     """
     ds_name = ds_repo_id.split("/")[-1]
-    artifact_dir = (
-        Path("tests/data/save_policy_to_safetensors") / f"{ds_name}_{policy_name}_{file_name_extra}"
-    )
+    artifact_dir = Path("tests/artifacts/policies") / f"{ds_name}_{policy_name}_{file_name_extra}"
     saved_output_dict = load_file(artifact_dir / "output_dict.safetensors")
     saved_grad_stats = load_file(artifact_dir / "grad_stats.safetensors")
     saved_param_stats = load_file(artifact_dir / "param_stats.safetensors")
diff --git a/tests/test_control_robot.py b/tests/robots/test_control_robot.py
similarity index 99%
rename from tests/test_control_robot.py
rename to tests/robots/test_control_robot.py
index cb54ae72..9b5ae52c 100644
--- a/tests/test_control_robot.py
+++ b/tests/robots/test_control_robot.py
@@ -51,7 +51,7 @@ from lerobot.configs.control import (
 )
 from lerobot.configs.policies import PreTrainedConfig
 from lerobot.scripts.control_robot import calibrate, record, replay, teleoperate
-from tests.test_robots import make_robot
+from tests.robots.test_robots import make_robot
 from tests.utils import TEST_ROBOT_TYPES, mock_calibration_dir, require_robot
 
 
diff --git a/tests/test_robots.py b/tests/robots/test_robots.py
similarity index 100%
rename from tests/test_robots.py
rename to tests/robots/test_robots.py
diff --git a/tests/test_io_utils.py b/tests/utils/test_io_utils.py
similarity index 100%
rename from tests/test_io_utils.py
rename to tests/utils/test_io_utils.py
diff --git a/tests/test_logging_utils.py b/tests/utils/test_logging_utils.py
similarity index 100%
rename from tests/test_logging_utils.py
rename to tests/utils/test_logging_utils.py
diff --git a/tests/test_random_utils.py b/tests/utils/test_random_utils.py
similarity index 100%
rename from tests/test_random_utils.py
rename to tests/utils/test_random_utils.py
diff --git a/tests/test_train_utils.py b/tests/utils/test_train_utils.py
similarity index 100%
rename from tests/test_train_utils.py
rename to tests/utils/test_train_utils.py