diff --git a/.github/workflows/ci.yaml b/.github/workflows/ci.yaml
index 9411948c..45ccf57f 100644
--- a/.github/workflows/ci.yaml
+++ b/.github/workflows/ci.yaml
@@ -159,6 +159,7 @@ jobs:
           - os: ubuntu-latest
             python-version: '3.8'  # torch 1.8.0 requires python >=3.6, <=3.8
             torch: '1.8.0'  # min torch version CI https://pypi.org/project/torchvision/
+          - os: gpu-latest  # do not pass python-version
     steps:
       - uses: actions/checkout@v3
       - uses: actions/setup-python@v4
@@ -176,6 +177,7 @@ jobs:
           fi
       - name: Check environment
         run: |
+          yolo checks
           echo "RUNNER_OS is ${{ runner.os }}"
           echo "GITHUB_EVENT_NAME is ${{ github.event_name }}"
           echo "GITHUB_WORKFLOW is ${{ github.workflow }}"
@@ -187,9 +189,14 @@ jobs:
           pip list
       - name: Pytest tests
         shell: bash  # for Windows compatibility
-        run: pytest --cov=ultralytics/ --cov-report xml tests/
+        run: |
+          if [ "${{ matrix.os }}" == "gpu-latest" ]; then
+              pytest --cov=ultralytics/ --cov-report xml tests/test_cuda.py
+          else
+              pytest --cov=ultralytics/ --cov-report xml tests/
+          fi
       - name: Upload Coverage Reports to CodeCov
-        if: github.repository == 'ultralytics/ultralytics' && matrix.os == 'ubuntu-latest' && matrix.python-version == '3.11'
+        if: github.repository == 'ultralytics/ultralytics' # && matrix.os == 'ubuntu-latest' && matrix.python-version == '3.11'
         uses: codecov/codecov-action@v3
         with:
           flags: Tests
diff --git a/.gitignore b/.gitignore
index d197c74d..c8987d84 100644
--- a/.gitignore
+++ b/.gitignore
@@ -140,6 +140,7 @@ dmypy.json
 datasets/
 runs/
 wandb/
+tests/
 .DS_Store
 
 # Neural Network weights -----------------------------------------------------------------------------------------------
@@ -158,6 +159,7 @@ weights/
 *_web_model/
 *_openvino_model/
 *_paddle_model/
+pnnx*
 
 # Autogenerated files for tests
 /ultralytics/assets/
diff --git a/docs/models/sam.md b/docs/models/sam.md
index d6ffd4d8..5b781cd2 100644
--- a/docs/models/sam.md
+++ b/docs/models/sam.md
@@ -51,7 +51,7 @@ The Segment Anything Model can be employed for a multitude of downstream tasks t
         model('ultralytics/assets/zidane.jpg', bboxes=[439, 437, 524, 709])
 
         # Run inference with points prompt
-        model.predict('ultralytics/assets/zidane.jpg', points=[900, 370], labels=[1])
+        model('ultralytics/assets/zidane.jpg', points=[900, 370], labels=[1])
         ```
 
 !!! example "Segment everything"
diff --git a/docs/reference/utils/checks.md b/docs/reference/utils/checks.md
index 96c45080..e4493562 100644
--- a/docs/reference/utils/checks.md
+++ b/docs/reference/utils/checks.md
@@ -80,3 +80,11 @@ keywords: Ultralytics, utility checks, ASCII, check_version, pip_update, check_p
 ---
 ## ::: ultralytics.utils.checks.print_args
 <br><br>
+
+---
+## ::: ultralytics.utils.checks.cuda_device_count
+<br><br>
+
+---
+## ::: ultralytics.utils.checks.cuda_is_available
+<br><br>
diff --git a/tests/conftest.py b/tests/conftest.py
index 235e2b77..50ea353f 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -44,5 +44,5 @@ def pytest_terminal_summary(terminalreporter, exitstatus, config):
         Path(file).unlink(missing_ok=True)
 
     # Remove directories
-    for directory in ['.pytest_cache/', TMP]:
+    for directory in [ROOT / '../.pytest_cache', TMP]:
         shutil.rmtree(directory, ignore_errors=True)
diff --git a/tests/test_cli.py b/tests/test_cli.py
index 2a97918e..51d2c123 100644
--- a/tests/test_cli.py
+++ b/tests/test_cli.py
@@ -63,7 +63,7 @@ def test_export(model, format):
 
 def test_rtdetr(task='detect', model='yolov8n-rtdetr.yaml', data='coco8.yaml'):
     # Warning: MUST use imgsz=640
-    run(f'yolo train {task} model={model} data={data} imgsz=640 epochs=1, cache = disk')  # add coma, space to args
+    run(f'yolo train {task} model={model} data={data} --imgsz= 640 epochs =1, cache = disk')  # add coma, spaces to args
     run(f"yolo predict {task} model={model} source={ASSETS / 'bus.jpg'} imgsz=640 save save_crop save_txt")
 
 
diff --git a/tests/test_cuda.py b/tests/test_cuda.py
new file mode 100644
index 00000000..7f893672
--- /dev/null
+++ b/tests/test_cuda.py
@@ -0,0 +1,73 @@
+# Ultralytics YOLO 🚀, AGPL-3.0 license
+import subprocess
+from pathlib import Path
+
+import pytest
+import torch
+
+from ultralytics import YOLO
+from ultralytics.utils import ASSETS, SETTINGS
+
+CUDA_IS_AVAILABLE = torch.cuda.is_available()
+CUDA_DEVICE_COUNT = torch.cuda.device_count()
+
+WEIGHTS_DIR = Path(SETTINGS['weights_dir'])
+MODEL = WEIGHTS_DIR / 'path with spaces' / 'yolov8n.pt'  # test spaces in path
+DATA = 'coco8.yaml'
+
+
+def test_checks():
+    from ultralytics.utils.checks import cuda_device_count, cuda_is_available
+
+    assert cuda_device_count() == CUDA_DEVICE_COUNT
+    assert cuda_is_available() == CUDA_IS_AVAILABLE
+
+
+@pytest.mark.skipif(not CUDA_IS_AVAILABLE, reason='CUDA is not available')
+def test_train():
+    YOLO(MODEL).train(data=DATA, imgsz=64, epochs=1, batch=-1, device=0)  # also test AutoBatch, requires imgsz>=64
+
+
+@pytest.mark.skipif(CUDA_DEVICE_COUNT < 2, reason=f'DDP is not available, {CUDA_DEVICE_COUNT} device(s) found')
+def test_train_ddp():
+    YOLO(MODEL).train(data=DATA, imgsz=64, epochs=1, device=[0, 1])  # requires imgsz>=64
+
+
+@pytest.mark.skipif(not CUDA_IS_AVAILABLE, reason='CUDA is not available')
+def test_utils_benchmarks():
+    from ultralytics.utils.benchmarks import ProfileModels
+
+    YOLO(MODEL).export(format='engine', imgsz=32, dynamic=True, batch=1)  # pre-export engine model, auto-device
+    ProfileModels([MODEL], imgsz=32, half=False, min_time=1, num_timed_runs=3, num_warmup_runs=1).profile()
+
+
+@pytest.mark.skipif(not CUDA_IS_AVAILABLE, reason='CUDA is not available')
+def test_predict_sam():
+    from ultralytics import SAM
+
+    # Load a model
+    model = SAM(WEIGHTS_DIR / 'sam_b.pt')
+
+    # Display model information (optional)
+    model.info()
+
+    # Run inference
+    model(ASSETS / 'bus.jpg', device=0)
+
+    # Run inference with bboxes prompt
+    model(ASSETS / 'zidane.jpg', bboxes=[439, 437, 524, 709], device=0)
+
+    # Run inference with points prompt
+    model(ASSETS / 'zidane.jpg', points=[900, 370], labels=[1], device=0)
+
+
+@pytest.mark.skipif(not CUDA_IS_AVAILABLE, reason='CUDA is not available')
+def test_model_tune():
+    subprocess.run('pip install ray[tune]'.split(), check=True)
+    YOLO('yolov8n-cls.yaml').tune(data='imagenet10',
+                                  grace_period=1,
+                                  max_samples=1,
+                                  imgsz=32,
+                                  epochs=1,
+                                  plots=False,
+                                  device='cpu')
diff --git a/tests/test_python.py b/tests/test_python.py
index 1fc5fc81..eb418cd7 100644
--- a/tests/test_python.py
+++ b/tests/test_python.py
@@ -1,5 +1,6 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
 
+import contextlib
 import shutil
 from copy import copy
 from pathlib import Path
@@ -38,6 +39,8 @@ def test_model_methods():
     model = model.load(MODEL)
     model.to('cpu')
     model.fuse()
+    model.clear_callback('on_train_start')
+    model._reset_callbacks()
 
     # Model properties
     _ = model.names
@@ -314,6 +317,15 @@ def test_events():
     events(cfg)
 
 
+def test_cfg_init():
+    from ultralytics.cfg import check_dict_alignment, copy_default_cfg, smart_value
+
+    with contextlib.suppress(SyntaxError):
+        check_dict_alignment({'a': 1}, {'b': 2})
+    copy_default_cfg()
+    [smart_value(x) for x in ['none', 'true', 'false']]
+
+
 def test_utils_init():
     from ultralytics.utils import get_git_branch, get_git_origin_url, get_ubuntu_version, is_github_actions_ci
 
@@ -354,6 +366,7 @@ def test_utils_torchutils():
     time_sync()
 
 
+@pytest.mark.skipif(not ONLINE, reason='environment is offline')
 def test_utils_downloads():
     from ultralytics.utils.downloads import get_google_drive_file_info
 
@@ -422,8 +435,11 @@ def test_nn_modules_block():
     BottleneckCSP(c1, c2)(x)
 
 
+@pytest.mark.skipif(not ONLINE, reason='environment is offline')
 def test_hub():
     from ultralytics.hub import export_fmts_hub, logout
+    from ultralytics.hub.utils import smart_request
 
     export_fmts_hub()
     logout()
+    smart_request('GET', 'http://github.com', progress=True)
diff --git a/ultralytics/__init__.py b/ultralytics/__init__.py
index 4686f988..190a63ee 100644
--- a/ultralytics/__init__.py
+++ b/ultralytics/__init__.py
@@ -1,6 +1,6 @@
 # Ultralytics YOLO 🚀, AGPL-3.0 license
 
-__version__ = '8.0.162'
+__version__ = '8.0.163'
 
 from ultralytics.models import RTDETR, SAM, YOLO
 from ultralytics.models.fastsam import FastSAM
diff --git a/ultralytics/cfg/__init__.py b/ultralytics/cfg/__init__.py
index 2b17340c..19558ec7 100644
--- a/ultralytics/cfg/__init__.py
+++ b/ultralytics/cfg/__init__.py
@@ -4,7 +4,6 @@ import contextlib
 import re
 import shutil
 import sys
-from difflib import get_close_matches
 from pathlib import Path
 from types import SimpleNamespace
 from typing import Dict, List, Union
@@ -177,6 +176,8 @@ def check_dict_alignment(base: Dict, custom: Dict, e=None):
     base_keys, custom_keys = (set(x.keys()) for x in (base, custom))
     mismatched = [k for k in custom_keys if k not in base_keys]
     if mismatched:
+        from difflib import get_close_matches
+
         string = ''
         for x in mismatched:
             matches = get_close_matches(x, base_keys)  # key list
@@ -373,11 +374,7 @@ def entrypoint(debug=''):
         mode = DEFAULT_CFG.mode or 'predict'
         LOGGER.warning(f"WARNING ⚠️ 'mode' is missing. Valid modes are {MODES}. Using default 'mode={mode}'.")
     elif mode not in MODES:
-        if mode not in ('checks', checks):
-            raise ValueError(f"Invalid 'mode={mode}'. Valid modes are {MODES}.\n{CLI_HELP_MSG}")
-        LOGGER.warning("WARNING ⚠️ 'yolo mode=checks' is deprecated. Use 'yolo checks' instead.")
-        checks.check_yolo()
-        return
+        raise ValueError(f"Invalid 'mode={mode}'. Valid modes are {MODES}.\n{CLI_HELP_MSG}")
 
     # Task
     task = overrides.pop('task', None)
diff --git a/ultralytics/data/augment.py b/ultralytics/data/augment.py
index 68160d44..d24f7982 100644
--- a/ultralytics/data/augment.py
+++ b/ultralytics/data/augment.py
@@ -483,7 +483,7 @@ class RandomHSV:
         self.vgain = vgain
 
     def __call__(self, labels):
-        """Applies random horizontal or vertical flip to an image with a given probability."""
+        """Applies image HSV augmentation"""
         img = labels['img']
         if self.hgain or self.sgain or self.vgain:
             r = np.random.uniform(-1, 1, 3) * [self.hgain, self.sgain, self.vgain] + 1  # random gains
@@ -501,6 +501,7 @@ class RandomHSV:
 
 
 class RandomFlip:
+    """Applies random horizontal or vertical flip to an image with a given probability."""
 
     def __init__(self, p=0.5, direction='horizontal', flip_idx=None) -> None:
         assert direction in ['horizontal', 'vertical'], f'Support direction `horizontal` or `vertical`, got {direction}'
@@ -643,7 +644,9 @@ class CopyPaste:
 
 
 class Albumentations:
-    """YOLOv8 Albumentations class (optional, only used if package is installed)"""
+    """Albumentations transformations. Optional, uninstall package to disable.
+    Applies Blur, Median Blur, convert to grayscale, Contrast Limited Adaptive Histogram Equalization,
+    random change of brightness and contrast, RandomGamma and lowering of image quality by compression."""
 
     def __init__(self, p=1.0):
         """Initialize the transform object for YOLO bbox formatted params."""
diff --git a/ultralytics/engine/exporter.py b/ultralytics/engine/exporter.py
index 4283f6a8..abe648e6 100644
--- a/ultralytics/engine/exporter.py
+++ b/ultralytics/engine/exporter.py
@@ -159,7 +159,10 @@ class Exporter:
             raise ValueError(f"Invalid export format='{format}'. Valid formats are {fmts}")
         jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, paddle, ncnn = flags  # export booleans
 
-        # Load PyTorch model
+        # Device
+        if format == 'engine' and self.args.device is None:
+            LOGGER.warning('WARNING ⚠️ TensorRT requires GPU export, automatically assigning device=0')
+            self.args.device = '0'
         self.device = select_device('cpu' if self.args.device is None else self.args.device)
 
         # Checks
diff --git a/ultralytics/utils/benchmarks.py b/ultralytics/utils/benchmarks.py
index 8113c2dd..d92d0625 100644
--- a/ultralytics/utils/benchmarks.py
+++ b/ultralytics/utils/benchmarks.py
@@ -182,6 +182,7 @@ class ProfileModels:
                  num_warmup_runs=10,
                  min_time=60,
                  imgsz=640,
+                 half=True,
                  trt=True,
                  device=None):
         self.paths = paths
@@ -189,6 +190,7 @@ class ProfileModels:
         self.num_warmup_runs = num_warmup_runs
         self.min_time = min_time
         self.imgsz = imgsz
+        self.half = half
         self.trt = trt  # run TensorRT profiling
         self.device = device or torch.device(0 if torch.cuda.is_available() else 'cpu')
 
@@ -209,12 +211,12 @@ class ProfileModels:
                 model_info = model.info()
                 if self.trt and self.device.type != 'cpu' and not engine_file.is_file():
                     engine_file = model.export(format='engine',
-                                               half=True,
+                                               half=self.half,
                                                imgsz=self.imgsz,
                                                device=self.device,
                                                verbose=False)
                 onnx_file = model.export(format='onnx',
-                                         half=True,
+                                         half=self.half,
                                          imgsz=self.imgsz,
                                          simplify=True,
                                          device=self.device,
diff --git a/ultralytics/utils/checks.py b/ultralytics/utils/checks.py
index aa48430b..d1de180b 100644
--- a/ultralytics/utils/checks.py
+++ b/ultralytics/utils/checks.py
@@ -507,3 +507,28 @@ def print_args(args: Optional[dict] = None, show_file=True, show_func=False):
         file = Path(file).stem
     s = (f'{file}: ' if show_file else '') + (f'{func}: ' if show_func else '')
     LOGGER.info(colorstr(s) + ', '.join(f'{k}={strip_auth(v)}' for k, v in args.items()))
+
+
+def cuda_device_count() -> int:
+    """Get the number of NVIDIA GPUs available in the environment.
+
+    Returns:
+        (int): The number of NVIDIA GPUs available.
+    """
+    try:
+        # Run the nvidia-smi command and capture its output
+        output = subprocess.check_output(['nvidia-smi', '--query-gpu=count', '--format=csv,noheader,nounits'],
+                                         encoding='utf-8')
+        return int(output.strip())
+    except (subprocess.CalledProcessError, FileNotFoundError):
+        # If the command fails or nvidia-smi is not found, assume no GPUs are available
+        return 0
+
+
+def cuda_is_available() -> bool:
+    """Check if CUDA is available in the environment.
+
+    Returns:
+        (bool): True if one or more NVIDIA GPUs are available, False otherwise.
+    """
+    return cuda_device_count() > 0