mlflow-modal-deploy 0.2.0__py3-none-any.whl

mlflow_modal/__init__.py

@@ -0,0 +1,29 @@
+"""
+MLflow Modal Deployment Plugin
+
+Deploy MLflow models to Modal's serverless infrastructure.
+See https://modal.com for more information.
+"""
+
+from mlflow_modal.deployment import (
+    DEFAULT_CPU,
+    DEFAULT_GPU,
+    DEFAULT_MEMORY,
+    DEFAULT_TIMEOUT,
+    SUPPORTED_GPUS,
+    ModalDeploymentClient,
+    run_local,
+    target_help,
+)
+
+__version__ = "0.1.0"
+__all__ = [
+    "ModalDeploymentClient",
+    "run_local",
+    "target_help",
+    "SUPPORTED_GPUS",
+    "DEFAULT_GPU",
+    "DEFAULT_MEMORY",
+    "DEFAULT_CPU",
+    "DEFAULT_TIMEOUT",
+]

mlflow_modal/deployment.py

@@ -0,0 +1,779 @@
+"""
+Modal Deployment Client for MLflow models.
+
+This module provides the ModalDeploymentClient class for deploying
+MLflow models to Modal's serverless infrastructure.
+"""
+
+import json
+import logging
+import os
+import re
+import subprocess
+import textwrap
+import urllib.parse
+from typing import Any
+
+import requests
+from mlflow.deployments import BaseDeploymentClient, PredictionsResponse
+from mlflow.exceptions import MlflowException
+from mlflow.models import Model
+from mlflow.protos.databricks_pb2 import INVALID_PARAMETER_VALUE, RESOURCE_DOES_NOT_EXIST
+from mlflow.pyfunc import FLAVOR_NAME as PYFUNC_FLAVOR_NAME
+from mlflow.tracking.artifact_utils import _download_artifact_from_uri
+from mlflow.utils.file_utils import TempDir
+
+_logger = logging.getLogger(__name__)
+
+# Default configuration values
+DEFAULT_GPU = None
+DEFAULT_MEMORY = 512  # MB
+DEFAULT_CPU = 1.0
+DEFAULT_TIMEOUT = 300  # seconds
+DEFAULT_CONTAINER_IDLE_TIMEOUT = 60  # seconds
+DEFAULT_ALLOW_CONCURRENT_INPUTS = 1
+DEFAULT_MIN_CONTAINERS = 0
+DEFAULT_MAX_CONTAINERS = None
+DEFAULT_SCALEDOWN_WINDOW = None
+
+# Supported GPU types
+SUPPORTED_GPUS = ["T4", "L4", "A10G", "A100", "A100-80GB", "H100"]
+
+
+def _get_model_requirements(model_path: str) -> tuple[list[str], list[str]]:
+    """
+    Extract Python requirements from an MLflow model.
+
+    Returns:
+        Tuple of (pip_requirements, wheel_files) where wheel_files are paths
+        to .whl files in the model's code/ directory.
+    """
+    requirements = []
+    wheel_files = []
+
+    # Check for wheel files in code/ directory
+    code_dir = os.path.join(model_path, "code")
+    if os.path.exists(code_dir):
+        for filename in os.listdir(code_dir):
+            if filename.endswith(".whl"):
+                wheel_files.append(os.path.join(code_dir, filename))
+                _logger.info(f"Found wheel dependency: {filename}")
+
+    req_file = os.path.join(model_path, "requirements.txt")
+    if os.path.exists(req_file):
+        with open(req_file) as f:
+            for line in f:
+                line = line.strip()
+                if line and not line.startswith("#") and not line.lower().startswith("mlflow"):
+                    # Skip wheel references - we handle them separately
+                    if not line.endswith(".whl") and "code/" not in line:
+                        requirements.append(line)
+        return requirements, wheel_files
+
+    conda_file = os.path.join(model_path, "conda.yaml")
+    if os.path.exists(conda_file):
+        try:
+            import yaml
+
+            with open(conda_file) as f:
+                conda_env = yaml.safe_load(f)
+            for dep in conda_env.get("dependencies", []):
+                if isinstance(dep, dict) and "pip" in dep:
+                    requirements.extend(
+                        pip_dep
+                        for pip_dep in dep["pip"]
+                        if not pip_dep.lower().startswith("mlflow") and not pip_dep.endswith(".whl")
+                    )
+                elif isinstance(dep, str) and not dep.startswith("python"):
+                    if not dep.lower().startswith("mlflow"):
+                        requirements.append(dep)
+        except Exception as e:
+            _logger.warning(f"Failed to parse conda.yaml: {e}")
+    return requirements, wheel_files
+
+
+def _get_model_python_version(model_path: str) -> str | None:
+    """Extract Python version from an MLflow model."""
+    conda_file = os.path.join(model_path, "conda.yaml")
+    if os.path.exists(conda_file):
+        try:
+            import yaml
+
+            with open(conda_file) as f:
+                conda_env = yaml.safe_load(f)
+            for dep in conda_env.get("dependencies", []):
+                if isinstance(dep, str) and dep.startswith("python"):
+                    version = dep.split("=")[-1].split(">")[-1].split("<")[-1]
+                    parts = version.split(".")
+                    if len(parts) >= 2:
+                        return f"{parts[0]}.{parts[1]}"
+        except Exception as e:
+            _logger.warning(f"Failed to parse Python version: {e}")
+    return None
+
+
+def _clear_volume(modal, volume_name: str) -> None:
+    """Clear Modal volume to allow redeployment."""
+    try:
+        volume = modal.Volume.from_name(volume_name)
+        for entry in volume.listdir("/"):
+            try:
+                volume.remove_file(f"/{entry.path}")
+            except Exception:
+                pass
+        _logger.info(f"Cleared volume: {volume_name}")
+    except Exception as e:
+        _logger.debug(f"Could not clear volume {volume_name}: {e}")
+
+
+def _get_preferred_deployment_flavor(model_config):
+    """Obtain the flavor MLflow prefers for deployment."""
+    if PYFUNC_FLAVOR_NAME in model_config.flavors:
+        return PYFUNC_FLAVOR_NAME
+    raise MlflowException(
+        message=(
+            "The specified model does not contain the python_function flavor "
+            "which is required for Modal deployment. "
+            f"The model contains the following flavors: {list(model_config.flavors.keys())}."
+        ),
+        error_code=RESOURCE_DOES_NOT_EXIST,
+    )
+
+
+def _validate_deployment_flavor(model_config, flavor):
+    """Validate that the specified flavor is supported and present in the model."""
+    if flavor != PYFUNC_FLAVOR_NAME:
+        raise MlflowException(
+            message=f"Flavor '{flavor}' is not supported for Modal deployment. "
+            f"Only '{PYFUNC_FLAVOR_NAME}' is supported.",
+            error_code=INVALID_PARAMETER_VALUE,
+        )
+
+    if flavor not in model_config.flavors:
+        raise MlflowException(
+            message=f"The specified model does not contain the '{flavor}' flavor. "
+            f"Available flavors: {list(model_config.flavors.keys())}.",
+            error_code=RESOURCE_DOES_NOT_EXIST,
+        )
+
+
+def _import_modal():
+    """Import modal and raise helpful error if not installed."""
+    try:
+        import modal
+
+        return modal
+    except ImportError as e:
+        raise MlflowException(
+            "The `modal` package is required for Modal deployments. Please install it with: pip install modal"
+        ) from e
+
+
+def _generate_modal_app_code(
+    app_name: str,
+    model_path: str,
+    config: dict[str, Any],
+    model_requirements: list[str] | None = None,
+    wheel_filenames: list[str] | None = None,
+) -> str:
+    """
+    Generate Modal app Python code for serving an MLflow model.
+
+    Args:
+        app_name: Name of the Modal app
+        model_path: Path to the MLflow model directory
+        config: Deployment configuration
+        model_requirements: List of pip requirements
+        wheel_filenames: List of wheel filenames (just names, not paths) to install from volume
+    """
+    gpu_config = config.get("gpu")
+    memory = config.get("memory", DEFAULT_MEMORY)
+    cpu = config.get("cpu", DEFAULT_CPU)
+    timeout = config.get("timeout", DEFAULT_TIMEOUT)
+    container_idle_timeout = config.get("container_idle_timeout", DEFAULT_CONTAINER_IDLE_TIMEOUT)
+    enable_batching = config.get("enable_batching", False)
+    max_batch_size = config.get("max_batch_size", 8)
+    batch_wait_ms = config.get("batch_wait_ms", 100)
+    python_version = config.get("python_version", "3.10")
+
+    gpu_str = f'"{gpu_config}"' if gpu_config else "None"
+
+    pip_packages = ["mlflow"]
+    if model_requirements:
+        pip_packages.extend(model_requirements)
+    pip_install_str = ", ".join(f'"{pkg}"' for pkg in pip_packages)
+
+    scaling_parts = []
+    min_containers = config.get("min_containers", DEFAULT_MIN_CONTAINERS)
+    max_containers = config.get("max_containers")
+    scaledown_window = config.get("scaledown_window")
+    allow_concurrent_inputs = config.get("allow_concurrent_inputs", DEFAULT_ALLOW_CONCURRENT_INPUTS)
+
+    if min_containers is not None and min_containers > 0:
+        scaling_parts.append(f"min_containers={min_containers}")
+    if max_containers is not None:
+        scaling_parts.append(f"max_containers={max_containers}")
+    if scaledown_window is not None:
+        scaling_parts.append(f"scaledown_window={scaledown_window}")
+    if allow_concurrent_inputs != DEFAULT_ALLOW_CONCURRENT_INPUTS:
+        scaling_parts.append(f"allow_concurrent_inputs={allow_concurrent_inputs}")
+
+    scaling_str = "\n            ".join(f"{part}," for part in scaling_parts)
+
+    # Generate wheel installation code if wheels are present
+    wheel_install_code = ""
+    if wheel_filenames:
+        wheel_paths = [f"/model/wheels/{whl}" for whl in wheel_filenames]
+        wheel_install_code = textwrap.dedent(f"""
+                # Install wheel dependencies from volume
+                import subprocess
+                import sys
+                wheel_files = {wheel_paths}
+                for whl in wheel_files:
+                    subprocess.check_call([sys.executable, "-m", "pip", "install", whl, "--quiet"])
+        """)
+
+    code = textwrap.dedent(f'''
+        """
+        Modal app for serving MLflow model: {app_name}
+        Auto-generated by mlflow-modal plugin
+        """
+        import modal
+        import os
+
+        app = modal.App("{app_name}")
+
+        model_volume = modal.Volume.from_name("{app_name}-model-volume", create_if_missing=True)
+        MODEL_DIR = "/model"
+
+        image = (
+            modal.Image.debian_slim(python_version="{python_version}")
+            .pip_install({pip_install_str})
+        )
+
+        @app.cls(
+            image=image,
+            gpu={gpu_str},
+            memory={memory},
+            cpu={cpu},
+            timeout={timeout},
+            container_idle_timeout={container_idle_timeout},
+            {scaling_str}
+            volumes={{MODEL_DIR: model_volume}},
+        )
+        class MLflowModel:
+            @modal.enter()
+            def load_model(self):
+                import mlflow.pyfunc
+                model_volume.reload()
+{wheel_install_code}
+                self.model = mlflow.pyfunc.load_model(MODEL_DIR)
+
+    ''')
+
+    if enable_batching:
+        code += textwrap.dedent(f"""
+            @modal.batched(max_batch_size={max_batch_size}, wait_ms={batch_wait_ms})
+            def predict_batch(self, inputs: list[dict]) -> list[dict]:
+                import pandas as pd
+                results = []
+                for input_data in inputs:
+                    df = pd.DataFrame(input_data)
+                    prediction = self.model.predict(df)
+                    results.append({{"predictions": prediction.tolist()}})
+                return results
+
+            @modal.web_endpoint(method="POST")
+            def predict(self, input_data: dict) -> dict:
+                return self.predict_batch.local([input_data])[0]
+        """)
+    else:
+        code += textwrap.dedent("""
+            @modal.web_endpoint(method="POST")
+            def predict(self, input_data: dict) -> dict:
+                import pandas as pd
+                df = pd.DataFrame(input_data)
+                prediction = self.model.predict(df)
+                return {"predictions": prediction.tolist()}
+        """)
+
+    return code
+
+
+class ModalDeploymentClient(BaseDeploymentClient):
+    """
+    Client for deploying MLflow models to Modal.
+
+    Modal is a serverless platform for running Python code in the cloud.
+    This client enables deploying MLflow models as Modal web endpoints.
+
+    Args:
+        target_uri: A URI that follows one of the following formats:
+            - ``modal``: Uses default Modal workspace from environment
+            - ``modal:/workspace-name``: Uses the specified workspace
+
+    Example:
+        .. code-block:: python
+
+            from mlflow.deployments import get_deploy_client
+
+            client = get_deploy_client("modal")
+            client.create_deployment(
+                name="my-model",
+                model_uri="runs:/abc123/model",
+                config={"gpu": "T4", "enable_batching": True},
+            )
+    """
+
+    def __init__(self, target_uri: str):
+        super().__init__(target_uri=target_uri)
+        self.workspace = self._parse_workspace_from_uri(target_uri)
+        self._validate_modal_auth()
+
+    def _parse_workspace_from_uri(self, target_uri: str) -> str | None:
+        """Parse workspace name from target URI."""
+        parsed = urllib.parse.urlparse(target_uri)
+        if parsed.scheme == "modal":
+            path = parsed.path.strip("/")
+            return path or None
+        return None
+
+    def _validate_modal_auth(self):
+        """Validate that Modal authentication is configured."""
+        _import_modal()
+
+    def _default_deployment_config(self) -> dict[str, Any]:
+        """Return default deployment configuration."""
+        return {
+            "gpu": DEFAULT_GPU,
+            "memory": DEFAULT_MEMORY,
+            "cpu": DEFAULT_CPU,
+            "timeout": DEFAULT_TIMEOUT,
+            "container_idle_timeout": DEFAULT_CONTAINER_IDLE_TIMEOUT,
+            "enable_batching": False,
+            "max_batch_size": 8,
+            "batch_wait_ms": 100,
+            "allow_concurrent_inputs": DEFAULT_ALLOW_CONCURRENT_INPUTS,
+            "min_containers": DEFAULT_MIN_CONTAINERS,
+            "max_containers": DEFAULT_MAX_CONTAINERS,
+            "scaledown_window": DEFAULT_SCALEDOWN_WINDOW,
+            "python_version": None,
+        }
+
+    def _apply_custom_config(self, config: dict[str, Any], custom_config: dict[str, Any] | None) -> dict[str, Any]:
+        """Apply custom configuration over defaults."""
+        if not custom_config:
+            return config
+
+        int_fields = {
+            "memory",
+            "timeout",
+            "container_idle_timeout",
+            "max_batch_size",
+            "batch_wait_ms",
+            "min_containers",
+            "max_containers",
+            "scaledown_window",
+            "allow_concurrent_inputs",
+        }
+        float_fields = {"cpu"}
+        bool_fields = {"enable_batching"}
+
+        for key, value in custom_config.items():
+            if key not in config:
+                config[key] = value
+                continue
+
+            if value is None:
+                config[key] = value
+            elif key in int_fields and not isinstance(value, int):
+                config[key] = int(value)
+            elif key in float_fields and not isinstance(value, float):
+                config[key] = float(value)
+            elif key in bool_fields and not isinstance(value, bool):
+                config[key] = str(value).lower() == "true"
+            else:
+                config[key] = value
+
+        if config.get("gpu") and config["gpu"] not in SUPPORTED_GPUS:
+            raise MlflowException(
+                f"Unsupported GPU type: {config['gpu']}. Supported types: {SUPPORTED_GPUS}",
+                error_code=INVALID_PARAMETER_VALUE,
+            )
+
+        return config
+
+    def create_deployment(
+        self,
+        name: str,
+        model_uri: str,
+        flavor: str | None = None,
+        config: dict[str, Any] | None = None,
+        endpoint: str | None = None,
+    ) -> dict[str, Any]:
+        """
+        Deploy an MLflow model to Modal.
+
+        Args:
+            name: Name of the deployment (will be used as Modal app name)
+            model_uri: URI of the MLflow model to deploy
+            flavor: Model flavor (only python_function supported)
+            config: Deployment configuration dict with keys:
+                - gpu: GPU type (T4, L4, A10G, A100, A100-80GB, H100)
+                - memory: Memory in MB (default: 512)
+                - cpu: CPU cores (default: 1.0)
+                - timeout: Request timeout in seconds (default: 300)
+                - container_idle_timeout: Idle timeout in seconds (default: 60)
+                - enable_batching: Enable dynamic batching (default: False)
+                - max_batch_size: Max batch size (default: 8)
+                - batch_wait_ms: Batch wait time in ms (default: 100)
+                - min_containers: Minimum containers (default: 0)
+                - max_containers: Maximum containers (default: None)
+                - python_version: Python version (default: auto-detect)
+            endpoint: Unused, kept for API compatibility
+
+        Returns:
+            Dictionary with deployment information including endpoint URL
+        """
+        modal = _import_modal()
+
+        with TempDir() as tmp_dir:
+            local_model_path = _download_artifact_from_uri(model_uri, output_path=tmp_dir.path())
+            model_config = Model.load(local_model_path)
+
+            if flavor is None:
+                flavor = _get_preferred_deployment_flavor(model_config)
+            else:
+                _validate_deployment_flavor(model_config, flavor)
+
+            deployment_config = self._default_deployment_config()
+            deployment_config = self._apply_custom_config(deployment_config, config)
+
+            if deployment_config.get("python_version") is None:
+                detected_version = _get_model_python_version(local_model_path)
+                deployment_config["python_version"] = detected_version or "3.10"
+
+            model_requirements, wheel_files = _get_model_requirements(local_model_path)
+            if model_requirements:
+                _logger.info(f"Detected pip requirements: {model_requirements}")
+            if wheel_files:
+                _logger.info(f"Detected wheel files: {[os.path.basename(w) for w in wheel_files]}")
+
+            wheel_filenames = [os.path.basename(w) for w in wheel_files] if wheel_files else None
+            app_code = _generate_modal_app_code(
+                name, local_model_path, deployment_config, model_requirements, wheel_filenames
+            )
+
+            app_file = os.path.join(tmp_dir.path(), "modal_app.py")
+            with open(app_file, "w") as f:
+                f.write(app_code)
+
+            volume_name = f"{name}-model-volume"
+            _clear_volume(modal, volume_name)
+
+            _logger.info(f"Uploading model to Modal volume: {volume_name}")
+            volume = modal.Volume.from_name(volume_name, create_if_missing=True)
+            volume.batch_upload(local_model_path, "/", force=True)
+
+            # Upload wheel files to a separate wheels/ directory in the volume
+            if wheel_files:
+                wheels_dir = os.path.join(tmp_dir.path(), "wheels")
+                os.makedirs(wheels_dir, exist_ok=True)
+                import shutil
+
+                for whl in wheel_files:
+                    shutil.copy(whl, wheels_dir)
+                _logger.info(f"Uploading {len(wheel_files)} wheel file(s) to volume")
+                volume.batch_upload(wheels_dir, "/wheels", force=True)
+
+            _logger.info(f"Deploying Modal app: {name}")
+            deploy_cmd = ["modal", "deploy", app_file]
+            if self.workspace:
+                deploy_cmd.extend(["--env", self.workspace])
+
+            result = subprocess.run(deploy_cmd, capture_output=True, text=True, cwd=tmp_dir.path())
+
+            # Log deployment output for debugging
+            if result.stdout:
+                for line in result.stdout.strip().split("\n"):
+                    _logger.info(f"[modal deploy] {line}")
+            if result.stderr:
+                for line in result.stderr.strip().split("\n"):
+                    if result.returncode != 0:
+                        _logger.error(f"[modal deploy] {line}")
+                    else:
+                        _logger.warning(f"[modal deploy] {line}")
+
+            if result.returncode != 0:
+                error_msg = result.stderr or result.stdout or "Unknown error"
+                raise MlflowException(
+                    f"Failed to deploy Modal app. Build logs:\n{error_msg}\n\n"
+                    "Tip: Run with MLFLOW_LOGGING_LEVEL=DEBUG for more details.",
+                    error_code=INVALID_PARAMETER_VALUE,
+                )
+
+            endpoint_url = None
+            for line in result.stdout.split("\n"):
+                if "https://" in line and ".modal.run" in line:
+                    match = re.search(r"(https://[^\s]+\.modal\.run[^\s]*)", line)
+                    if match:
+                        endpoint_url = match.group(1)
+                        break
+
+            return {
+                "name": name,
+                "flavor": flavor,
+                "model_uri": model_uri,
+                "endpoint_url": endpoint_url,
+                "config": deployment_config,
+            }
+
+    def update_deployment(
+        self,
+        name: str,
+        model_uri: str | None = None,
+        flavor: str | None = None,
+        config: dict[str, Any] | None = None,
+        endpoint: str | None = None,
+    ) -> dict[str, Any]:
+        """Update an existing Modal deployment."""
+        if model_uri is None:
+            raise MlflowException(
+                "model_uri is required for updating a Modal deployment",
+                error_code=INVALID_PARAMETER_VALUE,
+            )
+        return self.create_deployment(name, model_uri, flavor, config, endpoint)
+
+    def delete_deployment(self, name: str, endpoint: str | None = None) -> dict[str, Any]:
+        """Delete a Modal deployment."""
+        modal = _import_modal()
+
+        stop_cmd = ["modal", "app", "stop", name]
+        if self.workspace:
+            stop_cmd.extend(["--env", self.workspace])
+
+        result = subprocess.run(stop_cmd, capture_output=True, text=True)
+
+        volume_name = f"{name}-model-volume"
+        try:
+            volume = modal.Volume.from_name(volume_name)
+            volume.delete()
+            _logger.info(f"Deleted volume: {volume_name}")
+        except Exception as e:
+            _logger.debug(f"Could not delete volume {volume_name}: {e}")
+
+        if result.returncode != 0 and "not found" not in result.stderr.lower():
+            raise MlflowException(
+                f"Failed to delete Modal app: {result.stderr}",
+                error_code=INVALID_PARAMETER_VALUE,
+            )
+
+        return {"name": name, "deleted": True}
+
+    def list_deployments(self, endpoint: str | None = None) -> list[dict[str, Any]]:
+        """List all Modal deployments."""
+        list_cmd = ["modal", "app", "list", "--json"]
+        if self.workspace:
+            list_cmd.extend(["--env", self.workspace])
+
+        result = subprocess.run(list_cmd, capture_output=True, text=True)
+
+        if result.returncode != 0:
+            raise MlflowException(
+                f"Failed to list Modal apps: {result.stderr}",
+                error_code=INVALID_PARAMETER_VALUE,
+            )
+
+        try:
+            apps = json.loads(result.stdout) if result.stdout.strip() else []
+        except json.JSONDecodeError:
+            apps = []
+
+        return [{"name": app.get("name", app.get("app_id", "unknown"))} for app in apps]
+
+    def get_deployment(self, name: str, endpoint: str | None = None) -> dict[str, Any]:
+        """Get information about a specific Modal deployment."""
+        deployments = self.list_deployments()
+        for deployment in deployments:
+            if deployment.get("name") == name:
+                return deployment
+
+        raise MlflowException(
+            f"Deployment '{name}' not found",
+            error_code=RESOURCE_DOES_NOT_EXIST,
+        )
+
+    def predict(
+        self,
+        deployment_name: str | None = None,
+        inputs: Any = None,
+        endpoint: str | None = None,
+    ) -> PredictionsResponse:
+        """Make predictions using a deployed Modal model."""
+        if deployment_name is None:
+            raise MlflowException(
+                "deployment_name is required",
+                error_code=INVALID_PARAMETER_VALUE,
+            )
+
+        deployment = self.get_deployment(deployment_name)
+        endpoint_url = deployment.get("endpoint_url")
+
+        if not endpoint_url:
+            list_cmd = ["modal", "app", "list", "--json"]
+            if self.workspace:
+                list_cmd.extend(["--env", self.workspace])
+
+            result = subprocess.run(list_cmd, capture_output=True, text=True)
+            if result.returncode == 0:
+                try:
+                    apps = json.loads(result.stdout) if result.stdout.strip() else []
+                    for app in apps:
+                        if app.get("name") == deployment_name or app.get("app_id") == deployment_name:
+                            for func in app.get("functions", []):
+                                if "predict" in func.get("name", "").lower():
+                                    endpoint_url = func.get("web_url")
+                                    break
+                except json.JSONDecodeError:
+                    pass
+
+        if not endpoint_url:
+            raise MlflowException(
+                f"Could not find endpoint URL for deployment '{deployment_name}'. "
+                "The deployment may not have a web endpoint configured.",
+                error_code=RESOURCE_DOES_NOT_EXIST,
+            )
+
+        response = requests.post(endpoint_url, json=inputs, timeout=300)
+        response.raise_for_status()
+
+        return PredictionsResponse(predictions=response.json())
+
+
+def run_local(
+    target_uri: str,
+    name: str,
+    model_uri: str,
+    flavor: str | None = None,
+    config: dict[str, Any] | None = None,
+) -> None:
+    """
+    Run a Modal deployment locally using `modal serve`.
+
+    Args:
+        target_uri: Target URI (e.g., "modal")
+        name: Name for the local deployment
+        model_uri: URI of the MLflow model to deploy
+        flavor: Model flavor (only python_function supported)
+        config: Deployment configuration
+    """
+    _import_modal()
+
+    with TempDir() as tmp_dir:
+        local_model_path = _download_artifact_from_uri(model_uri, output_path=tmp_dir.path())
+
+        deployment_config = ModalDeploymentClient("modal")._default_deployment_config()
+        if config:
+            deployment_config.update(config)
+
+        if deployment_config.get("python_version") is None:
+            detected_version = _get_model_python_version(local_model_path)
+            deployment_config["python_version"] = detected_version or "3.10"
+
+        model_requirements, wheel_files = _get_model_requirements(local_model_path)
+        if model_requirements:
+            _logger.info(f"Detected pip requirements: {model_requirements}")
+        if wheel_files:
+            _logger.info(f"Detected wheel files: {[os.path.basename(w) for w in wheel_files]}")
+
+        wheel_filenames = [os.path.basename(w) for w in wheel_files] if wheel_files else None
+        app_code = _generate_modal_app_code(
+            name, local_model_path, deployment_config, model_requirements, wheel_filenames
+        )
+
+        app_file = os.path.join(tmp_dir.path(), "modal_app.py")
+        with open(app_file, "w") as f:
+            f.write(app_code)
+
+        _logger.info(f"Starting local Modal server for {name}...")
+        subprocess.run(["modal", "serve", app_file], cwd=tmp_dir.path())
+
+
+def target_help() -> str:
+    """Return help text for the Modal deployment target."""
+    return """
+    MLflow Modal Deployment Plugin
+    ==============================
+
+    Deploy MLflow models to Modal's serverless platform (https://modal.com).
+
+    Installation
+    ------------
+    pip install mlflow-modal
+
+    Target URI Format
+    -----------------
+    - ``modal``: Use default workspace from Modal authentication
+    - ``modal:/workspace-name``: Use a specific Modal environment/workspace
+
+    Authentication
+    --------------
+    Modal authentication is handled via the Modal CLI or environment variables:
+    - Run ``modal setup`` to configure authentication interactively
+    - Or set MODAL_TOKEN_ID and MODAL_TOKEN_SECRET environment variables
+
+    Configuration Options
+    ---------------------
+    Pass these options via the ``config`` parameter in create_deployment():
+
+    Resource Configuration:
+    - ``gpu``: GPU type (T4, L4, A10G, A100, A100-80GB, H100)
+    - ``memory``: Memory allocation in MB (default: 512)
+    - ``cpu``: CPU cores (default: 1.0)
+    - ``timeout``: Request timeout in seconds (default: 300)
+
+    Scaling Configuration:
+    - ``min_containers``: Minimum containers to keep warm (default: 0)
+    - ``max_containers``: Maximum containers to scale to (default: None)
+    - ``container_idle_timeout``: Container idle timeout in seconds (default: 60)
+    - ``scaledown_window``: Time window for scale-down decisions in seconds
+    - ``allow_concurrent_inputs``: Concurrent inputs per container (default: 1)
+
+    Batching Configuration:
+    - ``enable_batching``: Enable dynamic batching (default: False)
+    - ``max_batch_size``: Maximum batch size when batching enabled (default: 8)
+    - ``batch_wait_ms``: Batch wait time in milliseconds (default: 100)
+
+    Example
+    -------
+    .. code-block:: python
+
+        from mlflow.deployments import get_deploy_client
+
+        client = get_deploy_client("modal")
+
+        deployment = client.create_deployment(
+            name="my-classifier",
+            model_uri="runs:/abc123/model",
+            config={
+                "gpu": "T4",
+                "memory": 2048,
+                "min_containers": 1,
+                "enable_batching": True,
+            }
+        )
+
+        predictions = client.predict(
+            deployment_name="my-classifier",
+            inputs={"feature1": [1, 2, 3], "feature2": [4, 5, 6]}
+        )
+
+    CLI Usage
+    ---------
+    .. code-block:: bash
+
+        mlflow deployments create -t modal -m runs:/abc123/model --name my-model
+        mlflow deployments list -t modal
+        mlflow deployments get -t modal --name my-model
+        mlflow deployments delete -t modal --name my-model
+    """

mlflow_modal_deploy-0.2.0.dist-info/METADATA

@@ -0,0 +1,242 @@
+Metadata-Version: 2.4
+Name: mlflow-modal-deploy
+Version: 0.2.0
+Summary: MLflow deployment plugin for Modal serverless GPU infrastructure (actively maintained)
+Home-page: https://github.com/debu-sinha/mlflow-modal
+Author: Debu Sinha
+Author-email: debu.sinha@example.com
+License: Apache-2.0
+Project-URL: Homepage, https://github.com/debu-sinha/mlflow-modal-deploy
+Project-URL: Repository, https://github.com/debu-sinha/mlflow-modal-deploy
+Project-URL: Issues, https://github.com/debu-sinha/mlflow-modal-deploy/issues
+Keywords: mlflow,modal,deployment,serverless,machine-learning
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: mlflow>=2.10.0
+Requires-Dist: modal>=0.64.0
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0.0; extra == "dev"
+Requires-Dist: pytest-cov>=4.0.0; extra == "dev"
+Requires-Dist: ruff>=0.1.0; extra == "dev"
+Requires-Dist: twine>=4.0.0; extra == "dev"
+Requires-Dist: build>=1.0.0; extra == "dev"
+Dynamic: author-email
+Dynamic: home-page
+Dynamic: license-file
+Dynamic: requires-python
+
+# mlflow-modal-deploy
+
+[![CI](https://github.com/debu-sinha/mlflow-modal-deploy/actions/workflows/ci.yml/badge.svg)](https://github.com/debu-sinha/mlflow-modal-deploy/actions/workflows/ci.yml)
+[![PyPI version](https://badge.fury.io/py/mlflow-modal-deploy.svg)](https://badge.fury.io/py/mlflow-modal-deploy)
+[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
+[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
+
+Deploy MLflow models to [Modal](https://modal.com)'s serverless GPU infrastructure with a single command.
+
+## Installation
+
+```bash
+pip install mlflow-modal-deploy
+```
+
+## Features
+
+- **One-command deployment**: Deploy any MLflow model to Modal's serverless infrastructure
+- **GPU support**: T4, L4, A10G, A100, A100-80GB, H100
+- **Auto-scaling**: Configure min/max containers, scale-down windows
+- **Dynamic batching**: Built-in request batching for high-throughput workloads
+- **Automatic dependency detection**: Extracts requirements from model artifacts
+- **Wheel file support**: Handles private dependencies packaged as wheel files
+- **MLflow CLI integration**: Use familiar `mlflow deployments` commands
+
+## Quick Start
+
+### Python API
+
+```python
+from mlflow.deployments import get_deploy_client
+
+# Get the Modal deployment client
+client = get_deploy_client("modal")
+
+# Deploy a model
+deployment = client.create_deployment(
+    name="my-classifier",
+    model_uri="runs:/abc123/model",
+    config={
+        "gpu": "T4",
+        "memory": 2048,
+        "min_containers": 1,
+    }
+)
+
+print(f"Deployed to: {deployment['endpoint_url']}")
+
+# Make predictions
+predictions = client.predict(
+    deployment_name="my-classifier",
+    inputs={"feature1": [1, 2, 3], "feature2": [4, 5, 6]}
+)
+```
+
+### CLI
+
+```bash
+# Deploy a model
+mlflow deployments create -t modal -m runs:/abc123/model --name my-model
+
+# Deploy with GPU
+mlflow deployments create -t modal -m runs:/abc123/model --name gpu-model \
+    -C gpu=T4 -C memory=4096
+
+# List deployments
+mlflow deployments list -t modal
+
+# Get deployment info
+mlflow deployments get -t modal --name my-model
+
+# Delete deployment
+mlflow deployments delete -t modal --name my-model
+```
+
+## Configuration Options
+
+| Option | Type | Default | Description |
+|--------|------|---------|-------------|
+| `gpu` | str | None | GPU type: T4, L4, A10G, A100, A100-80GB, H100 |
+| `memory` | int | 512 | Memory allocation in MB |
+| `cpu` | float | 1.0 | CPU cores |
+| `timeout` | int | 300 | Request timeout in seconds |
+| `container_idle_timeout` | int | 60 | Container idle timeout in seconds |
+| `min_containers` | int | 0 | Minimum warm containers |
+| `max_containers` | int | None | Maximum containers |
+| `enable_batching` | bool | False | Enable dynamic batching |
+| `max_batch_size` | int | 8 | Max batch size when batching enabled |
+| `batch_wait_ms` | int | 100 | Batch wait time in milliseconds |
+| `python_version` | str | auto | Python version (auto-detected from model) |
+
+## Authentication
+
+Configure Modal authentication before deploying:
+
+```bash
+# Interactive setup
+modal setup
+
+# Or use environment variables
+export MODAL_TOKEN_ID=your-token-id
+export MODAL_TOKEN_SECRET=your-token-secret
+```
+
+## Advanced Usage
+
+### Deploy to Specific Workspace
+
+```python
+# Use workspace-specific URI
+client = get_deploy_client("modal:/production")
+```
+
+Or via CLI:
+
+```bash
+mlflow deployments create -t modal:/production -m runs:/abc123/model --name my-model
+```
+
+### High-Throughput Deployment with Batching
+
+```python
+client.create_deployment(
+    name="batch-classifier",
+    model_uri="runs:/abc123/model",
+    config={
+        "gpu": "A100",
+        "enable_batching": True,
+        "max_batch_size": 32,
+        "batch_wait_ms": 50,
+        "min_containers": 2,
+        "max_containers": 20,
+    }
+)
+```
+
+### Models with Private Dependencies
+
+If your model includes wheel files in the `code/` directory, they are automatically detected and installed:
+
+```
+model/
+├── MLmodel
+├── requirements.txt
+├── code/
+│   └── my_private_package-1.0.0-py3-none-any.whl  # Auto-detected
+└── ...
+```
+
+### Local Development
+
+Test your deployment locally before deploying to Modal:
+
+```python
+from mlflow_modal import run_local
+
+run_local(
+    target_uri="modal",
+    name="test-model",
+    model_uri="runs:/abc123/model",
+    config={"gpu": "T4"}
+)
+```
+
+## Requirements
+
+- Python 3.10+
+- MLflow 2.10.0+
+- Modal 0.64.0+
+
+## Contributing
+
+Contributions welcome! Please see [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
+
+### Development Setup
+
+```bash
+# Clone the repository
+git clone https://github.com/debu-sinha/mlflow-modal-deploy.git
+cd mlflow-modal-deploy
+
+# Install with dev dependencies
+uv sync --extra dev
+
+# Install pre-commit hooks
+uv run pre-commit install
+
+# Run tests
+uv run pytest tests/ -v
+```
+
+## License
+
+Apache License 2.0
+
+## Acknowledgments
+
+- [MLflow](https://mlflow.org/) - Open source platform for the ML lifecycle
+- [Modal](https://modal.com/) - Serverless cloud for AI/ML
+
+## Support
+
+- [GitHub Issues](https://github.com/debu-sinha/mlflow-modal-deploy/issues) - Bug reports and feature requests
+- [MLflow Slack](https://mlflow.org/slack) - Community discussion

mlflow_modal_deploy-0.2.0.dist-info/RECORD

@@ -0,0 +1,8 @@
+mlflow_modal/__init__.py,sha256=AirbpCl81V1qpEt8t6vP2sdMrxCmhWCiXZw9Xc24f_E,545
+mlflow_modal/deployment.py,sha256=iameFi9sc_BJVmMU6BNYAmgdFk1qlS_1Ulk9V42Zyl0,29484
+mlflow_modal_deploy-0.2.0.dist-info/licenses/LICENSE,sha256=2rWZc3jxeOoY5vVVgr7oI-oJJoTBNat6vchfwzPK8pQ,10759
+mlflow_modal_deploy-0.2.0.dist-info/METADATA,sha256=gnnNqAEP4wQNUKG22YuEb1h5DcpbLcGXhvkjlEFizEU,6848
+mlflow_modal_deploy-0.2.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+mlflow_modal_deploy-0.2.0.dist-info/entry_points.txt,sha256=EzJMKb71JNQlHMnU__zFqfNhQh6w8XEuGko54SSroow,75
+mlflow_modal_deploy-0.2.0.dist-info/top_level.txt,sha256=pZaQEqSmi6kuM89dAA953A3qb6DswJRUmWgOfK4OyH0,13
+mlflow_modal_deploy-0.2.0.dist-info/RECORD,,

mlflow_modal_deploy-0.2.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (80.9.0)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

mlflow_modal_deploy-0.2.0.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[mlflow.deployments]
+modal = mlflow_modal.deployment:ModalDeploymentClient

mlflow_modal_deploy-0.2.0.dist-info/licenses/LICENSE

@@ -0,0 +1,190 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to the Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   Copyright 2025 Debu Sinha
+
+   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.

mlflow_modal_deploy-0.2.0.dist-info/top_level.txt

@@ -0,0 +1 @@
+mlflow_modal