pulumiverse-scaleway 1.23.0a1736837529__py3-none-any.whl

pulumiverse_scaleway/inference_deployment.py

@@ -0,0 +1,822 @@
+# coding=utf-8
+# *** WARNING: this file was generated by the Pulumi Terraform Bridge (tfgen) Tool. ***
+# *** Do not edit by hand unless you're certain you know what you are doing! ***
+
+import copy
+import warnings
+import sys
+import pulumi
+import pulumi.runtime
+from typing import Any, Mapping, Optional, Sequence, Union, overload
+if sys.version_info >= (3, 11):
+    from typing import NotRequired, TypedDict, TypeAlias
+else:
+    from typing_extensions import NotRequired, TypedDict, TypeAlias
+from . import _utilities
+from . import outputs
+from ._inputs import *
+
+__all__ = ['InferenceDeploymentArgs', 'InferenceDeployment']
+
+@pulumi.input_type
+class InferenceDeploymentArgs:
+    def __init__(__self__, *,
+                 model_name: pulumi.Input[str],
+                 node_type: pulumi.Input[str],
+                 accept_eula: Optional[pulumi.Input[bool]] = None,
+                 max_size: Optional[pulumi.Input[int]] = None,
+                 min_size: Optional[pulumi.Input[int]] = None,
+                 name: Optional[pulumi.Input[str]] = None,
+                 private_endpoint: Optional[pulumi.Input['InferenceDeploymentPrivateEndpointArgs']] = None,
+                 project_id: Optional[pulumi.Input[str]] = None,
+                 public_endpoint: Optional[pulumi.Input['InferenceDeploymentPublicEndpointArgs']] = None,
+                 region: Optional[pulumi.Input[str]] = None,
+                 tags: Optional[pulumi.Input[Sequence[pulumi.Input[str]]]] = None):
+        """
+        The set of arguments for constructing a InferenceDeployment resource.
+        :param pulumi.Input[str] model_name: The model name to use for the deployment. Model names can be found in Console or using Scaleway's CLI (`scw inference model list`)
+        :param pulumi.Input[str] node_type: The node type to use for the deployment. Node types can be found using Scaleway's CLI (`scw inference node-type list`)
+        :param pulumi.Input[bool] accept_eula: Some models (e.g Meta Llama) require end-user license agreements. Set `true` to accept.
+        :param pulumi.Input[int] max_size: The maximum size of the pool.
+        :param pulumi.Input[int] min_size: The minimum size of the pool.
+        :param pulumi.Input[str] name: The deployment name.
+        :param pulumi.Input['InferenceDeploymentPrivateEndpointArgs'] private_endpoint: Configuration of the deployment's private endpoint.
+        :param pulumi.Input[str] project_id: `project_id`) The ID of the project the deployment is associated with.
+        :param pulumi.Input['InferenceDeploymentPublicEndpointArgs'] public_endpoint: Configuration of the deployment's public endpoint.
+        :param pulumi.Input[str] region: `region`) The region in which the deployment is created.
+        :param pulumi.Input[Sequence[pulumi.Input[str]]] tags: The tags associated with the deployment.
+        """
+        pulumi.set(__self__, "model_name", model_name)
+        pulumi.set(__self__, "node_type", node_type)
+        if accept_eula is not None:
+            pulumi.set(__self__, "accept_eula", accept_eula)
+        if max_size is not None:
+            pulumi.set(__self__, "max_size", max_size)
+        if min_size is not None:
+            pulumi.set(__self__, "min_size", min_size)
+        if name is not None:
+            pulumi.set(__self__, "name", name)
+        if private_endpoint is not None:
+            pulumi.set(__self__, "private_endpoint", private_endpoint)
+        if project_id is not None:
+            pulumi.set(__self__, "project_id", project_id)
+        if public_endpoint is not None:
+            pulumi.set(__self__, "public_endpoint", public_endpoint)
+        if region is not None:
+            pulumi.set(__self__, "region", region)
+        if tags is not None:
+            pulumi.set(__self__, "tags", tags)
+
+    @property
+    @pulumi.getter(name="modelName")
+    def model_name(self) -> pulumi.Input[str]:
+        """
+        The model name to use for the deployment. Model names can be found in Console or using Scaleway's CLI (`scw inference model list`)
+        """
+        return pulumi.get(self, "model_name")
+
+    @model_name.setter
+    def model_name(self, value: pulumi.Input[str]):
+        pulumi.set(self, "model_name", value)
+
+    @property
+    @pulumi.getter(name="nodeType")
+    def node_type(self) -> pulumi.Input[str]:
+        """
+        The node type to use for the deployment. Node types can be found using Scaleway's CLI (`scw inference node-type list`)
+        """
+        return pulumi.get(self, "node_type")
+
+    @node_type.setter
+    def node_type(self, value: pulumi.Input[str]):
+        pulumi.set(self, "node_type", value)
+
+    @property
+    @pulumi.getter(name="acceptEula")
+    def accept_eula(self) -> Optional[pulumi.Input[bool]]:
+        """
+        Some models (e.g Meta Llama) require end-user license agreements. Set `true` to accept.
+        """
+        return pulumi.get(self, "accept_eula")
+
+    @accept_eula.setter
+    def accept_eula(self, value: Optional[pulumi.Input[bool]]):
+        pulumi.set(self, "accept_eula", value)
+
+    @property
+    @pulumi.getter(name="maxSize")
+    def max_size(self) -> Optional[pulumi.Input[int]]:
+        """
+        The maximum size of the pool.
+        """
+        return pulumi.get(self, "max_size")
+
+    @max_size.setter
+    def max_size(self, value: Optional[pulumi.Input[int]]):
+        pulumi.set(self, "max_size", value)
+
+    @property
+    @pulumi.getter(name="minSize")
+    def min_size(self) -> Optional[pulumi.Input[int]]:
+        """
+        The minimum size of the pool.
+        """
+        return pulumi.get(self, "min_size")
+
+    @min_size.setter
+    def min_size(self, value: Optional[pulumi.Input[int]]):
+        pulumi.set(self, "min_size", value)
+
+    @property
+    @pulumi.getter
+    def name(self) -> Optional[pulumi.Input[str]]:
+        """
+        The deployment name.
+        """
+        return pulumi.get(self, "name")
+
+    @name.setter
+    def name(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "name", value)
+
+    @property
+    @pulumi.getter(name="privateEndpoint")
+    def private_endpoint(self) -> Optional[pulumi.Input['InferenceDeploymentPrivateEndpointArgs']]:
+        """
+        Configuration of the deployment's private endpoint.
+        """
+        return pulumi.get(self, "private_endpoint")
+
+    @private_endpoint.setter
+    def private_endpoint(self, value: Optional[pulumi.Input['InferenceDeploymentPrivateEndpointArgs']]):
+        pulumi.set(self, "private_endpoint", value)
+
+    @property
+    @pulumi.getter(name="projectId")
+    def project_id(self) -> Optional[pulumi.Input[str]]:
+        """
+        `project_id`) The ID of the project the deployment is associated with.
+        """
+        return pulumi.get(self, "project_id")
+
+    @project_id.setter
+    def project_id(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "project_id", value)
+
+    @property
+    @pulumi.getter(name="publicEndpoint")
+    def public_endpoint(self) -> Optional[pulumi.Input['InferenceDeploymentPublicEndpointArgs']]:
+        """
+        Configuration of the deployment's public endpoint.
+        """
+        return pulumi.get(self, "public_endpoint")
+
+    @public_endpoint.setter
+    def public_endpoint(self, value: Optional[pulumi.Input['InferenceDeploymentPublicEndpointArgs']]):
+        pulumi.set(self, "public_endpoint", value)
+
+    @property
+    @pulumi.getter
+    def region(self) -> Optional[pulumi.Input[str]]:
+        """
+        `region`) The region in which the deployment is created.
+        """
+        return pulumi.get(self, "region")
+
+    @region.setter
+    def region(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "region", value)
+
+    @property
+    @pulumi.getter
+    def tags(self) -> Optional[pulumi.Input[Sequence[pulumi.Input[str]]]]:
+        """
+        The tags associated with the deployment.
+        """
+        return pulumi.get(self, "tags")
+
+    @tags.setter
+    def tags(self, value: Optional[pulumi.Input[Sequence[pulumi.Input[str]]]]):
+        pulumi.set(self, "tags", value)
+
+
+@pulumi.input_type
+class _InferenceDeploymentState:
+    def __init__(__self__, *,
+                 accept_eula: Optional[pulumi.Input[bool]] = None,
+                 created_at: Optional[pulumi.Input[str]] = None,
+                 max_size: Optional[pulumi.Input[int]] = None,
+                 min_size: Optional[pulumi.Input[int]] = None,
+                 model_id: Optional[pulumi.Input[str]] = None,
+                 model_name: Optional[pulumi.Input[str]] = None,
+                 name: Optional[pulumi.Input[str]] = None,
+                 node_type: Optional[pulumi.Input[str]] = None,
+                 private_endpoint: Optional[pulumi.Input['InferenceDeploymentPrivateEndpointArgs']] = None,
+                 project_id: Optional[pulumi.Input[str]] = None,
+                 public_endpoint: Optional[pulumi.Input['InferenceDeploymentPublicEndpointArgs']] = None,
+                 region: Optional[pulumi.Input[str]] = None,
+                 size: Optional[pulumi.Input[int]] = None,
+                 status: Optional[pulumi.Input[str]] = None,
+                 tags: Optional[pulumi.Input[Sequence[pulumi.Input[str]]]] = None,
+                 updated_at: Optional[pulumi.Input[str]] = None):
+        """
+        Input properties used for looking up and filtering InferenceDeployment resources.
+        :param pulumi.Input[bool] accept_eula: Some models (e.g Meta Llama) require end-user license agreements. Set `true` to accept.
+        :param pulumi.Input[str] created_at: The date and time of the creation of the deployment.
+        :param pulumi.Input[int] max_size: The maximum size of the pool.
+        :param pulumi.Input[int] min_size: The minimum size of the pool.
+        :param pulumi.Input[str] model_id: The model id used for the deployment.
+        :param pulumi.Input[str] model_name: The model name to use for the deployment. Model names can be found in Console or using Scaleway's CLI (`scw inference model list`)
+        :param pulumi.Input[str] name: The deployment name.
+        :param pulumi.Input[str] node_type: The node type to use for the deployment. Node types can be found using Scaleway's CLI (`scw inference node-type list`)
+        :param pulumi.Input['InferenceDeploymentPrivateEndpointArgs'] private_endpoint: Configuration of the deployment's private endpoint.
+        :param pulumi.Input[str] project_id: `project_id`) The ID of the project the deployment is associated with.
+        :param pulumi.Input['InferenceDeploymentPublicEndpointArgs'] public_endpoint: Configuration of the deployment's public endpoint.
+        :param pulumi.Input[str] region: `region`) The region in which the deployment is created.
+        :param pulumi.Input[int] size: The size of the pool.
+        :param pulumi.Input[str] status: The status of the deployment.
+        :param pulumi.Input[Sequence[pulumi.Input[str]]] tags: The tags associated with the deployment.
+        :param pulumi.Input[str] updated_at: The date and time of the last update of the deployment.
+        """
+        if accept_eula is not None:
+            pulumi.set(__self__, "accept_eula", accept_eula)
+        if created_at is not None:
+            pulumi.set(__self__, "created_at", created_at)
+        if max_size is not None:
+            pulumi.set(__self__, "max_size", max_size)
+        if min_size is not None:
+            pulumi.set(__self__, "min_size", min_size)
+        if model_id is not None:
+            pulumi.set(__self__, "model_id", model_id)
+        if model_name is not None:
+            pulumi.set(__self__, "model_name", model_name)
+        if name is not None:
+            pulumi.set(__self__, "name", name)
+        if node_type is not None:
+            pulumi.set(__self__, "node_type", node_type)
+        if private_endpoint is not None:
+            pulumi.set(__self__, "private_endpoint", private_endpoint)
+        if project_id is not None:
+            pulumi.set(__self__, "project_id", project_id)
+        if public_endpoint is not None:
+            pulumi.set(__self__, "public_endpoint", public_endpoint)
+        if region is not None:
+            pulumi.set(__self__, "region", region)
+        if size is not None:
+            pulumi.set(__self__, "size", size)
+        if status is not None:
+            pulumi.set(__self__, "status", status)
+        if tags is not None:
+            pulumi.set(__self__, "tags", tags)
+        if updated_at is not None:
+            pulumi.set(__self__, "updated_at", updated_at)
+
+    @property
+    @pulumi.getter(name="acceptEula")
+    def accept_eula(self) -> Optional[pulumi.Input[bool]]:
+        """
+        Some models (e.g Meta Llama) require end-user license agreements. Set `true` to accept.
+        """
+        return pulumi.get(self, "accept_eula")
+
+    @accept_eula.setter
+    def accept_eula(self, value: Optional[pulumi.Input[bool]]):
+        pulumi.set(self, "accept_eula", value)
+
+    @property
+    @pulumi.getter(name="createdAt")
+    def created_at(self) -> Optional[pulumi.Input[str]]:
+        """
+        The date and time of the creation of the deployment.
+        """
+        return pulumi.get(self, "created_at")
+
+    @created_at.setter
+    def created_at(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "created_at", value)
+
+    @property
+    @pulumi.getter(name="maxSize")
+    def max_size(self) -> Optional[pulumi.Input[int]]:
+        """
+        The maximum size of the pool.
+        """
+        return pulumi.get(self, "max_size")
+
+    @max_size.setter
+    def max_size(self, value: Optional[pulumi.Input[int]]):
+        pulumi.set(self, "max_size", value)
+
+    @property
+    @pulumi.getter(name="minSize")
+    def min_size(self) -> Optional[pulumi.Input[int]]:
+        """
+        The minimum size of the pool.
+        """
+        return pulumi.get(self, "min_size")
+
+    @min_size.setter
+    def min_size(self, value: Optional[pulumi.Input[int]]):
+        pulumi.set(self, "min_size", value)
+
+    @property
+    @pulumi.getter(name="modelId")
+    def model_id(self) -> Optional[pulumi.Input[str]]:
+        """
+        The model id used for the deployment.
+        """
+        return pulumi.get(self, "model_id")
+
+    @model_id.setter
+    def model_id(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "model_id", value)
+
+    @property
+    @pulumi.getter(name="modelName")
+    def model_name(self) -> Optional[pulumi.Input[str]]:
+        """
+        The model name to use for the deployment. Model names can be found in Console or using Scaleway's CLI (`scw inference model list`)
+        """
+        return pulumi.get(self, "model_name")
+
+    @model_name.setter
+    def model_name(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "model_name", value)
+
+    @property
+    @pulumi.getter
+    def name(self) -> Optional[pulumi.Input[str]]:
+        """
+        The deployment name.
+        """
+        return pulumi.get(self, "name")
+
+    @name.setter
+    def name(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "name", value)
+
+    @property
+    @pulumi.getter(name="nodeType")
+    def node_type(self) -> Optional[pulumi.Input[str]]:
+        """
+        The node type to use for the deployment. Node types can be found using Scaleway's CLI (`scw inference node-type list`)
+        """
+        return pulumi.get(self, "node_type")
+
+    @node_type.setter
+    def node_type(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "node_type", value)
+
+    @property
+    @pulumi.getter(name="privateEndpoint")
+    def private_endpoint(self) -> Optional[pulumi.Input['InferenceDeploymentPrivateEndpointArgs']]:
+        """
+        Configuration of the deployment's private endpoint.
+        """
+        return pulumi.get(self, "private_endpoint")
+
+    @private_endpoint.setter
+    def private_endpoint(self, value: Optional[pulumi.Input['InferenceDeploymentPrivateEndpointArgs']]):
+        pulumi.set(self, "private_endpoint", value)
+
+    @property
+    @pulumi.getter(name="projectId")
+    def project_id(self) -> Optional[pulumi.Input[str]]:
+        """
+        `project_id`) The ID of the project the deployment is associated with.
+        """
+        return pulumi.get(self, "project_id")
+
+    @project_id.setter
+    def project_id(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "project_id", value)
+
+    @property
+    @pulumi.getter(name="publicEndpoint")
+    def public_endpoint(self) -> Optional[pulumi.Input['InferenceDeploymentPublicEndpointArgs']]:
+        """
+        Configuration of the deployment's public endpoint.
+        """
+        return pulumi.get(self, "public_endpoint")
+
+    @public_endpoint.setter
+    def public_endpoint(self, value: Optional[pulumi.Input['InferenceDeploymentPublicEndpointArgs']]):
+        pulumi.set(self, "public_endpoint", value)
+
+    @property
+    @pulumi.getter
+    def region(self) -> Optional[pulumi.Input[str]]:
+        """
+        `region`) The region in which the deployment is created.
+        """
+        return pulumi.get(self, "region")
+
+    @region.setter
+    def region(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "region", value)
+
+    @property
+    @pulumi.getter
+    def size(self) -> Optional[pulumi.Input[int]]:
+        """
+        The size of the pool.
+        """
+        return pulumi.get(self, "size")
+
+    @size.setter
+    def size(self, value: Optional[pulumi.Input[int]]):
+        pulumi.set(self, "size", value)
+
+    @property
+    @pulumi.getter
+    def status(self) -> Optional[pulumi.Input[str]]:
+        """
+        The status of the deployment.
+        """
+        return pulumi.get(self, "status")
+
+    @status.setter
+    def status(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "status", value)
+
+    @property
+    @pulumi.getter
+    def tags(self) -> Optional[pulumi.Input[Sequence[pulumi.Input[str]]]]:
+        """
+        The tags associated with the deployment.
+        """
+        return pulumi.get(self, "tags")
+
+    @tags.setter
+    def tags(self, value: Optional[pulumi.Input[Sequence[pulumi.Input[str]]]]):
+        pulumi.set(self, "tags", value)
+
+    @property
+    @pulumi.getter(name="updatedAt")
+    def updated_at(self) -> Optional[pulumi.Input[str]]:
+        """
+        The date and time of the last update of the deployment.
+        """
+        return pulumi.get(self, "updated_at")
+
+    @updated_at.setter
+    def updated_at(self, value: Optional[pulumi.Input[str]]):
+        pulumi.set(self, "updated_at", value)
+
+
+class InferenceDeployment(pulumi.CustomResource):
+    @overload
+    def __init__(__self__,
+                 resource_name: str,
+                 opts: Optional[pulumi.ResourceOptions] = None,
+                 accept_eula: Optional[pulumi.Input[bool]] = None,
+                 max_size: Optional[pulumi.Input[int]] = None,
+                 min_size: Optional[pulumi.Input[int]] = None,
+                 model_name: Optional[pulumi.Input[str]] = None,
+                 name: Optional[pulumi.Input[str]] = None,
+                 node_type: Optional[pulumi.Input[str]] = None,
+                 private_endpoint: Optional[pulumi.Input[Union['InferenceDeploymentPrivateEndpointArgs', 'InferenceDeploymentPrivateEndpointArgsDict']]] = None,
+                 project_id: Optional[pulumi.Input[str]] = None,
+                 public_endpoint: Optional[pulumi.Input[Union['InferenceDeploymentPublicEndpointArgs', 'InferenceDeploymentPublicEndpointArgsDict']]] = None,
+                 region: Optional[pulumi.Input[str]] = None,
+                 tags: Optional[pulumi.Input[Sequence[pulumi.Input[str]]]] = None,
+                 __props__=None):
+        """
+        Creates and manages Scaleway Managed Inference deployments.
+        For more information, see [the documentation](https://www.scaleway.com/en/developers/api/inference/).
+
+        ## Example Usage
+
+        ### Basic
+
+        ```python
+        import pulumi
+        import pulumiverse_scaleway as scaleway
+
+        deployment = scaleway.InferenceDeployment("deployment",
+            name="tf-inference-deployment",
+            node_type="L4",
+            model_name="meta/llama-3.1-8b-instruct:fp8",
+            public_endpoint={
+                "is_enabled": True,
+            },
+            accept_eula=True)
+        ```
+
+        ## Import
+
+        Functions can be imported using, `{region}/{id}`, as shown below:
+
+        bash
+
+        ```sh
+        $ pulumi import scaleway:index/inferenceDeployment:InferenceDeployment deployment fr-par/11111111-1111-1111-1111-111111111111
+        ```
+
+        :param str resource_name: The name of the resource.
+        :param pulumi.ResourceOptions opts: Options for the resource.
+        :param pulumi.Input[bool] accept_eula: Some models (e.g Meta Llama) require end-user license agreements. Set `true` to accept.
+        :param pulumi.Input[int] max_size: The maximum size of the pool.
+        :param pulumi.Input[int] min_size: The minimum size of the pool.
+        :param pulumi.Input[str] model_name: The model name to use for the deployment. Model names can be found in Console or using Scaleway's CLI (`scw inference model list`)
+        :param pulumi.Input[str] name: The deployment name.
+        :param pulumi.Input[str] node_type: The node type to use for the deployment. Node types can be found using Scaleway's CLI (`scw inference node-type list`)
+        :param pulumi.Input[Union['InferenceDeploymentPrivateEndpointArgs', 'InferenceDeploymentPrivateEndpointArgsDict']] private_endpoint: Configuration of the deployment's private endpoint.
+        :param pulumi.Input[str] project_id: `project_id`) The ID of the project the deployment is associated with.
+        :param pulumi.Input[Union['InferenceDeploymentPublicEndpointArgs', 'InferenceDeploymentPublicEndpointArgsDict']] public_endpoint: Configuration of the deployment's public endpoint.
+        :param pulumi.Input[str] region: `region`) The region in which the deployment is created.
+        :param pulumi.Input[Sequence[pulumi.Input[str]]] tags: The tags associated with the deployment.
+        """
+        ...
+    @overload
+    def __init__(__self__,
+                 resource_name: str,
+                 args: InferenceDeploymentArgs,
+                 opts: Optional[pulumi.ResourceOptions] = None):
+        """
+        Creates and manages Scaleway Managed Inference deployments.
+        For more information, see [the documentation](https://www.scaleway.com/en/developers/api/inference/).
+
+        ## Example Usage
+
+        ### Basic
+
+        ```python
+        import pulumi
+        import pulumiverse_scaleway as scaleway
+
+        deployment = scaleway.InferenceDeployment("deployment",
+            name="tf-inference-deployment",
+            node_type="L4",
+            model_name="meta/llama-3.1-8b-instruct:fp8",
+            public_endpoint={
+                "is_enabled": True,
+            },
+            accept_eula=True)
+        ```
+
+        ## Import
+
+        Functions can be imported using, `{region}/{id}`, as shown below:
+
+        bash
+
+        ```sh
+        $ pulumi import scaleway:index/inferenceDeployment:InferenceDeployment deployment fr-par/11111111-1111-1111-1111-111111111111
+        ```
+
+        :param str resource_name: The name of the resource.
+        :param InferenceDeploymentArgs args: The arguments to use to populate this resource's properties.
+        :param pulumi.ResourceOptions opts: Options for the resource.
+        """
+        ...
+    def __init__(__self__, resource_name: str, *args, **kwargs):
+        resource_args, opts = _utilities.get_resource_args_opts(InferenceDeploymentArgs, pulumi.ResourceOptions, *args, **kwargs)
+        if resource_args is not None:
+            __self__._internal_init(resource_name, opts, **resource_args.__dict__)
+        else:
+            __self__._internal_init(resource_name, *args, **kwargs)
+
+    def _internal_init(__self__,
+                 resource_name: str,
+                 opts: Optional[pulumi.ResourceOptions] = None,
+                 accept_eula: Optional[pulumi.Input[bool]] = None,
+                 max_size: Optional[pulumi.Input[int]] = None,
+                 min_size: Optional[pulumi.Input[int]] = None,
+                 model_name: Optional[pulumi.Input[str]] = None,
+                 name: Optional[pulumi.Input[str]] = None,
+                 node_type: Optional[pulumi.Input[str]] = None,
+                 private_endpoint: Optional[pulumi.Input[Union['InferenceDeploymentPrivateEndpointArgs', 'InferenceDeploymentPrivateEndpointArgsDict']]] = None,
+                 project_id: Optional[pulumi.Input[str]] = None,
+                 public_endpoint: Optional[pulumi.Input[Union['InferenceDeploymentPublicEndpointArgs', 'InferenceDeploymentPublicEndpointArgsDict']]] = None,
+                 region: Optional[pulumi.Input[str]] = None,
+                 tags: Optional[pulumi.Input[Sequence[pulumi.Input[str]]]] = None,
+                 __props__=None):
+        opts = pulumi.ResourceOptions.merge(_utilities.get_resource_opts_defaults(), opts)
+        if not isinstance(opts, pulumi.ResourceOptions):
+            raise TypeError('Expected resource options to be a ResourceOptions instance')
+        if opts.id is None:
+            if __props__ is not None:
+                raise TypeError('__props__ is only valid when passed in combination with a valid opts.id to get an existing resource')
+            __props__ = InferenceDeploymentArgs.__new__(InferenceDeploymentArgs)
+
+            __props__.__dict__["accept_eula"] = accept_eula
+            __props__.__dict__["max_size"] = max_size
+            __props__.__dict__["min_size"] = min_size
+            if model_name is None and not opts.urn:
+                raise TypeError("Missing required property 'model_name'")
+            __props__.__dict__["model_name"] = model_name
+            __props__.__dict__["name"] = name
+            if node_type is None and not opts.urn:
+                raise TypeError("Missing required property 'node_type'")
+            __props__.__dict__["node_type"] = node_type
+            __props__.__dict__["private_endpoint"] = private_endpoint
+            __props__.__dict__["project_id"] = project_id
+            __props__.__dict__["public_endpoint"] = public_endpoint
+            __props__.__dict__["region"] = region
+            __props__.__dict__["tags"] = tags
+            __props__.__dict__["created_at"] = None
+            __props__.__dict__["model_id"] = None
+            __props__.__dict__["size"] = None
+            __props__.__dict__["status"] = None
+            __props__.__dict__["updated_at"] = None
+        super(InferenceDeployment, __self__).__init__(
+            'scaleway:index/inferenceDeployment:InferenceDeployment',
+            resource_name,
+            __props__,
+            opts)
+
+    @staticmethod
+    def get(resource_name: str,
+            id: pulumi.Input[str],
+            opts: Optional[pulumi.ResourceOptions] = None,
+            accept_eula: Optional[pulumi.Input[bool]] = None,
+            created_at: Optional[pulumi.Input[str]] = None,
+            max_size: Optional[pulumi.Input[int]] = None,
+            min_size: Optional[pulumi.Input[int]] = None,
+            model_id: Optional[pulumi.Input[str]] = None,
+            model_name: Optional[pulumi.Input[str]] = None,
+            name: Optional[pulumi.Input[str]] = None,
+            node_type: Optional[pulumi.Input[str]] = None,
+            private_endpoint: Optional[pulumi.Input[Union['InferenceDeploymentPrivateEndpointArgs', 'InferenceDeploymentPrivateEndpointArgsDict']]] = None,
+            project_id: Optional[pulumi.Input[str]] = None,
+            public_endpoint: Optional[pulumi.Input[Union['InferenceDeploymentPublicEndpointArgs', 'InferenceDeploymentPublicEndpointArgsDict']]] = None,
+            region: Optional[pulumi.Input[str]] = None,
+            size: Optional[pulumi.Input[int]] = None,
+            status: Optional[pulumi.Input[str]] = None,
+            tags: Optional[pulumi.Input[Sequence[pulumi.Input[str]]]] = None,
+            updated_at: Optional[pulumi.Input[str]] = None) -> 'InferenceDeployment':
+        """
+        Get an existing InferenceDeployment resource's state with the given name, id, and optional extra
+        properties used to qualify the lookup.
+
+        :param str resource_name: The unique name of the resulting resource.
+        :param pulumi.Input[str] id: The unique provider ID of the resource to lookup.
+        :param pulumi.ResourceOptions opts: Options for the resource.
+        :param pulumi.Input[bool] accept_eula: Some models (e.g Meta Llama) require end-user license agreements. Set `true` to accept.
+        :param pulumi.Input[str] created_at: The date and time of the creation of the deployment.
+        :param pulumi.Input[int] max_size: The maximum size of the pool.
+        :param pulumi.Input[int] min_size: The minimum size of the pool.
+        :param pulumi.Input[str] model_id: The model id used for the deployment.
+        :param pulumi.Input[str] model_name: The model name to use for the deployment. Model names can be found in Console or using Scaleway's CLI (`scw inference model list`)
+        :param pulumi.Input[str] name: The deployment name.
+        :param pulumi.Input[str] node_type: The node type to use for the deployment. Node types can be found using Scaleway's CLI (`scw inference node-type list`)
+        :param pulumi.Input[Union['InferenceDeploymentPrivateEndpointArgs', 'InferenceDeploymentPrivateEndpointArgsDict']] private_endpoint: Configuration of the deployment's private endpoint.
+        :param pulumi.Input[str] project_id: `project_id`) The ID of the project the deployment is associated with.
+        :param pulumi.Input[Union['InferenceDeploymentPublicEndpointArgs', 'InferenceDeploymentPublicEndpointArgsDict']] public_endpoint: Configuration of the deployment's public endpoint.
+        :param pulumi.Input[str] region: `region`) The region in which the deployment is created.
+        :param pulumi.Input[int] size: The size of the pool.
+        :param pulumi.Input[str] status: The status of the deployment.
+        :param pulumi.Input[Sequence[pulumi.Input[str]]] tags: The tags associated with the deployment.
+        :param pulumi.Input[str] updated_at: The date and time of the last update of the deployment.
+        """
+        opts = pulumi.ResourceOptions.merge(opts, pulumi.ResourceOptions(id=id))
+
+        __props__ = _InferenceDeploymentState.__new__(_InferenceDeploymentState)
+
+        __props__.__dict__["accept_eula"] = accept_eula
+        __props__.__dict__["created_at"] = created_at
+        __props__.__dict__["max_size"] = max_size
+        __props__.__dict__["min_size"] = min_size
+        __props__.__dict__["model_id"] = model_id
+        __props__.__dict__["model_name"] = model_name
+        __props__.__dict__["name"] = name
+        __props__.__dict__["node_type"] = node_type
+        __props__.__dict__["private_endpoint"] = private_endpoint
+        __props__.__dict__["project_id"] = project_id
+        __props__.__dict__["public_endpoint"] = public_endpoint
+        __props__.__dict__["region"] = region
+        __props__.__dict__["size"] = size
+        __props__.__dict__["status"] = status
+        __props__.__dict__["tags"] = tags
+        __props__.__dict__["updated_at"] = updated_at
+        return InferenceDeployment(resource_name, opts=opts, __props__=__props__)
+
+    @property
+    @pulumi.getter(name="acceptEula")
+    def accept_eula(self) -> pulumi.Output[Optional[bool]]:
+        """
+        Some models (e.g Meta Llama) require end-user license agreements. Set `true` to accept.
+        """
+        return pulumi.get(self, "accept_eula")
+
+    @property
+    @pulumi.getter(name="createdAt")
+    def created_at(self) -> pulumi.Output[str]:
+        """
+        The date and time of the creation of the deployment.
+        """
+        return pulumi.get(self, "created_at")
+
+    @property
+    @pulumi.getter(name="maxSize")
+    def max_size(self) -> pulumi.Output[int]:
+        """
+        The maximum size of the pool.
+        """
+        return pulumi.get(self, "max_size")
+
+    @property
+    @pulumi.getter(name="minSize")
+    def min_size(self) -> pulumi.Output[int]:
+        """
+        The minimum size of the pool.
+        """
+        return pulumi.get(self, "min_size")
+
+    @property
+    @pulumi.getter(name="modelId")
+    def model_id(self) -> pulumi.Output[str]:
+        """
+        The model id used for the deployment.
+        """
+        return pulumi.get(self, "model_id")
+
+    @property
+    @pulumi.getter(name="modelName")
+    def model_name(self) -> pulumi.Output[str]:
+        """
+        The model name to use for the deployment. Model names can be found in Console or using Scaleway's CLI (`scw inference model list`)
+        """
+        return pulumi.get(self, "model_name")
+
+    @property
+    @pulumi.getter
+    def name(self) -> pulumi.Output[str]:
+        """
+        The deployment name.
+        """
+        return pulumi.get(self, "name")
+
+    @property
+    @pulumi.getter(name="nodeType")
+    def node_type(self) -> pulumi.Output[str]:
+        """
+        The node type to use for the deployment. Node types can be found using Scaleway's CLI (`scw inference node-type list`)
+        """
+        return pulumi.get(self, "node_type")
+
+    @property
+    @pulumi.getter(name="privateEndpoint")
+    def private_endpoint(self) -> pulumi.Output[Optional['outputs.InferenceDeploymentPrivateEndpoint']]:
+        """
+        Configuration of the deployment's private endpoint.
+        """
+        return pulumi.get(self, "private_endpoint")
+
+    @property
+    @pulumi.getter(name="projectId")
+    def project_id(self) -> pulumi.Output[str]:
+        """
+        `project_id`) The ID of the project the deployment is associated with.
+        """
+        return pulumi.get(self, "project_id")
+
+    @property
+    @pulumi.getter(name="publicEndpoint")
+    def public_endpoint(self) -> pulumi.Output[Optional['outputs.InferenceDeploymentPublicEndpoint']]:
+        """
+        Configuration of the deployment's public endpoint.
+        """
+        return pulumi.get(self, "public_endpoint")
+
+    @property
+    @pulumi.getter
+    def region(self) -> pulumi.Output[str]:
+        """
+        `region`) The region in which the deployment is created.
+        """
+        return pulumi.get(self, "region")
+
+    @property
+    @pulumi.getter
+    def size(self) -> pulumi.Output[int]:
+        """
+        The size of the pool.
+        """
+        return pulumi.get(self, "size")
+
+    @property
+    @pulumi.getter
+    def status(self) -> pulumi.Output[str]:
+        """
+        The status of the deployment.
+        """
+        return pulumi.get(self, "status")
+
+    @property
+    @pulumi.getter
+    def tags(self) -> pulumi.Output[Optional[Sequence[str]]]:
+        """
+        The tags associated with the deployment.
+        """
+        return pulumi.get(self, "tags")
+
+    @property
+    @pulumi.getter(name="updatedAt")
+    def updated_at(self) -> pulumi.Output[str]:
+        """
+        The date and time of the last update of the deployment.
+        """
+        return pulumi.get(self, "updated_at")
+