diff --git a/.gitignore b/.gitignore
index f7b771bc73..cf3390782f 100644
--- a/.gitignore
+++ b/.gitignore
@@ -233,3 +233,6 @@ configs/development.yaml
 Dockerfile.dev
 discussion
 tmp_test
+
+# Auto-generated version file (created by setuptools_scm during build)
+vllm_omni/_version.py
diff --git a/pyproject.toml b/pyproject.toml
index 706d0152e1..8e9f9bf8b6 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -8,7 +8,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "vllm-omni"
-version = "0.14.0"
+dynamic = ["version", "dependencies"]
 description = "A framework for efficient model inference with omni-modality models"
 readme = "README.md"
 requires-python = ">=3.10,<3.14"
@@ -17,7 +17,6 @@ authors = [
     {name = "vLLM-Omni Team"}
 ]
 keywords = ["vllm", "multimodal", "diffusion", "transformer", "inference", "serving"]
-dynamic = ["dependencies"]
 classifiers = [
     "Development Status :: 3 - Alpha",
     "Intended Audience :: Developers",
@@ -85,6 +84,10 @@ include = ["vllm_omni*"]
 [tool.setuptools.package-data]
 "vllm_omni.model_executor.stage_configs" = ["*.yaml"]
 
+[tool.setuptools_scm]
+# Enable setuptools_scm for automatic version generation from git tags.
+# Version is written to vllm_omni/_version.py during build.
+
 [tool.ruff]
 line-length = 120
 exclude = [
diff --git a/setup.py b/setup.py
index e33c848a3f..496b032639 100644
--- a/setup.py
+++ b/setup.py
@@ -12,6 +12,7 @@
 from pathlib import Path
 
 from setuptools import setup
+from setuptools_scm import get_version
 
 
 def uninstall_onnxruntime() -> None:
@@ -104,6 +105,71 @@ def detect_target_device() -> str:
         return "cuda"
 
 
+def get_vllm_omni_version() -> str:
+    """
+    Get the vLLM-Omni version with device-specific suffix.
+
+    Version format: {base_version}+{device}
+    Examples:
+        - 0.14.0+cuda (release version with CUDA)
+        - 0.14.1.dev23+g1a2b3c4.rocm (dev version with ROCm)
+        - 0.15.0+npu (release version with NPU)
+
+    Environment variables:
+        VLLM_OMNI_VERSION_OVERRIDE: Override version completely
+        VLLM_OMNI_TARGET_DEVICE: Override device detection
+
+    Returns:
+        Version string with device suffix
+    """
+    # Allow complete version override via environment variable
+    if env_version := os.getenv("VLLM_OMNI_VERSION_OVERRIDE"):
+        print(f"Overriding vLLM-Omni version with {env_version} from VLLM_OMNI_VERSION_OVERRIDE")
+        os.environ["SETUPTOOLS_SCM_PRETEND_VERSION"] = env_version
+        # Get version without device suffix for override case
+        version = get_version()
+    else:
+        # Generate version from git tags via setuptools_scm (without writing yet)
+        try:
+            version = get_version()
+        except Exception as e:
+            print(f"Warning: Failed to get version from git, using fallback: {e}")
+            version = "dev"
+
+    # Determine separator: '+' for normal versions, '.' for dev versions with '+'
+    sep = "+" if "+" not in version else "."
+
+    # Append device-specific suffix
+    device = detect_target_device()
+
+    if device == "cuda":
+        version += f"{sep}cuda"
+    elif device == "rocm":
+        version += f"{sep}rocm"
+    elif device == "npu":
+        version += f"{sep}npu"
+    elif device == "xpu":
+        version += f"{sep}xpu"
+    elif device == "cpu":
+        version += f"{sep}cpu"
+    else:
+        raise RuntimeError(f"Unknown target device: {device}")
+
+    # Write the FULL version (with device suffix) to _version.py
+    # This ensures runtime import matches the package metadata version
+    version_file = Path(__file__).parent / "vllm_omni" / "_version.py"
+    version_file.parent.mkdir(parents=True, exist_ok=True)
+
+    with open(version_file, "w") as f:
+        f.write(f"""# Automatically generated by setuptools_scm
+__version__ = {version!r}
+__version_tuple__ = tuple(__version__.split('.'))
+""")
+
+    print(f"Generated version: {version}")
+    return version
+
+
 def load_requirements(file_path: Path) -> list[str]:
     """
     Load requirements from a file, supporting -r directive for recursive loading.
@@ -167,5 +233,6 @@ def get_install_requires() -> list[str]:
 
     # Setup configuration
     setup(
+        version=get_vllm_omni_version(),
         install_requires=install_requires,
     )
diff --git a/vllm_omni/version.py b/vllm_omni/version.py
index 15253fe6fd..e5f0b6b661 100644
--- a/vllm_omni/version.py
+++ b/vllm_omni/version.py
@@ -1,3 +1,25 @@
-__version__ = "0.14.0"
-__version_tuple__ = (0, 14, 0)
-# TODO: add auto version generation
+"""
+Version information for vLLM-Omni.
+
+The version is automatically generated from git tags via setuptools_scm
+and written to _version.py during package build.
+"""
+
+try:
+    # Import auto-generated version from _version.py (created by setuptools_scm)
+    from ._version import __version__, __version_tuple__
+except ImportError as e:
+    import warnings
+
+    warnings.warn(
+        f"Failed to import version from _version.py: {e}\n"
+        "This typically happens in development mode before building.\n"
+        "Using fallback version 'dev'.",
+        RuntimeWarning,
+        stacklevel=2,
+    )
+
+    __version__ = "dev"
+    __version_tuple__ = (0, 0, "dev")
+
+__all__ = ["__version__", "__version_tuple__"]