vellum-ai 0.14.63__py3-none-any.whl → 1.11.2__py3-none-any.whl

vellum/client/__init__.py

@@ -1,86 +1,69 @@
 # This file was auto-generated by Fern from our API Definition.
 
+import os
 import typing
-from .environment import VellumEnvironment
+
 import httpx
-from .core.client_wrapper import SyncClientWrapper
-from .resources.ad_hoc.client import AdHocClient
-from .resources.container_images.client import ContainerImagesClient
-from .resources.deployments.client import DeploymentsClient
-from .resources.document_indexes.client import DocumentIndexesClient
-from .resources.documents.client import DocumentsClient
-from .resources.folder_entities.client import FolderEntitiesClient
-from .resources.metric_definitions.client import MetricDefinitionsClient
-from .resources.ml_models.client import MlModelsClient
-from .resources.organizations.client import OrganizationsClient
-from .resources.prompts.client import PromptsClient
-from .resources.sandboxes.client import SandboxesClient
-from .resources.test_suite_runs.client import TestSuiteRunsClient
-from .resources.test_suites.client import TestSuitesClient
-from .resources.workflow_deployments.client import WorkflowDeploymentsClient
-from .resources.release_reviews.client import ReleaseReviewsClient
-from .resources.workflow_sandboxes.client import WorkflowSandboxesClient
-from .resources.workflows.client import WorkflowsClient
-from .resources.workspace_secrets.client import WorkspaceSecretsClient
-from .resources.workspaces.client import WorkspacesClient
-from .types.method_enum import MethodEnum
-from .types.execute_api_request_body import ExecuteApiRequestBody
-from .types.execute_api_request_headers_value import ExecuteApiRequestHeadersValue
-from .types.execute_api_request_bearer_token import ExecuteApiRequestBearerToken
-from .core.request_options import RequestOptions
-from .types.execute_api_response import ExecuteApiResponse
-from .core.serialization import convert_and_respect_annotation_metadata
-from .core.pydantic_utilities import parse_obj_as
-from json.decoder import JSONDecodeError
 from .core.api_error import ApiError
+from .core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
+from .core.request_options import RequestOptions
+from .environment import VellumEnvironment
+from .raw_client import AsyncRawVellum, RawVellum
+from .resources.ad_hoc.client import AdHocClient, AsyncAdHocClient
+from .resources.container_images.client import AsyncContainerImagesClient, ContainerImagesClient
+from .resources.deployments.client import AsyncDeploymentsClient, DeploymentsClient
+from .resources.document_indexes.client import AsyncDocumentIndexesClient, DocumentIndexesClient
+from .resources.documents.client import AsyncDocumentsClient, DocumentsClient
+from .resources.environments.client import AsyncEnvironmentsClient, EnvironmentsClient
+from .resources.events.client import AsyncEventsClient, EventsClient
+from .resources.folder_entities.client import AsyncFolderEntitiesClient, FolderEntitiesClient
+from .resources.integration_auth_configs.client import AsyncIntegrationAuthConfigsClient, IntegrationAuthConfigsClient
+from .resources.integration_providers.client import AsyncIntegrationProvidersClient, IntegrationProvidersClient
+from .resources.integrations.client import AsyncIntegrationsClient, IntegrationsClient
+from .resources.metric_definitions.client import AsyncMetricDefinitionsClient, MetricDefinitionsClient
+from .resources.ml_models.client import AsyncMlModelsClient, MlModelsClient
+from .resources.organizations.client import AsyncOrganizationsClient, OrganizationsClient
+from .resources.prompts.client import AsyncPromptsClient, PromptsClient
+from .resources.sandboxes.client import AsyncSandboxesClient, SandboxesClient
+from .resources.test_suite_runs.client import AsyncTestSuiteRunsClient, TestSuiteRunsClient
+from .resources.test_suites.client import AsyncTestSuitesClient, TestSuitesClient
+from .resources.uploaded_files.client import AsyncUploadedFilesClient, UploadedFilesClient
+from .resources.workflow_deployments.client import AsyncWorkflowDeploymentsClient, WorkflowDeploymentsClient
+from .resources.workflow_executions.client import AsyncWorkflowExecutionsClient, WorkflowExecutionsClient
+from .resources.workflow_sandboxes.client import AsyncWorkflowSandboxesClient, WorkflowSandboxesClient
+from .resources.workflows.client import AsyncWorkflowsClient, WorkflowsClient
+from .resources.workspace_secrets.client import AsyncWorkspaceSecretsClient, WorkspaceSecretsClient
+from .resources.workspaces.client import AsyncWorkspacesClient, WorkspacesClient
+from .types.api_version_enum import ApiVersionEnum
+from .types.code_execution_package import CodeExecutionPackage
 from .types.code_execution_runtime import CodeExecutionRuntime
 from .types.code_executor_input import CodeExecutorInput
-from .types.code_execution_package import CodeExecutionPackage
-from .types.vellum_variable_type import VellumVariableType
 from .types.code_executor_response import CodeExecutorResponse
-from .errors.bad_request_error import BadRequestError
-from .types.prompt_deployment_input_request import PromptDeploymentInputRequest
-from .types.prompt_deployment_expand_meta_request import PromptDeploymentExpandMetaRequest
-from .types.raw_prompt_execution_overrides_request import RawPromptExecutionOverridesRequest
-from .types.execute_prompt_response import ExecutePromptResponse
-from .errors.forbidden_error import ForbiddenError
-from .errors.not_found_error import NotFoundError
-from .errors.internal_server_error import InternalServerError
+from .types.execute_api_request_bearer_token import ExecuteApiRequestBearerToken
+from .types.execute_api_request_body import ExecuteApiRequestBody
+from .types.execute_api_request_headers_value import ExecuteApiRequestHeadersValue
+from .types.execute_api_response import ExecuteApiResponse
 from .types.execute_prompt_event import ExecutePromptEvent
-import json
-from .types.workflow_request_input_request import WorkflowRequestInputRequest
-from .types.workflow_expand_meta_request import WorkflowExpandMetaRequest
+from .types.execute_prompt_response import ExecutePromptResponse
+from .types.execute_workflow_async_response import ExecuteWorkflowAsyncResponse
 from .types.execute_workflow_response import ExecuteWorkflowResponse
-from .types.workflow_execution_event_type import WorkflowExecutionEventType
-from .types.workflow_stream_event import WorkflowStreamEvent
-from .types.generate_request import GenerateRequest
 from .types.generate_options_request import GenerateOptionsRequest
+from .types.generate_request import GenerateRequest
 from .types.generate_response import GenerateResponse
 from .types.generate_stream_response import GenerateStreamResponse
+from .types.method_enum import MethodEnum
+from .types.prompt_deployment_expand_meta_request import PromptDeploymentExpandMetaRequest
+from .types.prompt_deployment_input_request import PromptDeploymentInputRequest
+from .types.raw_prompt_execution_overrides_request import RawPromptExecutionOverridesRequest
 from .types.search_request_options_request import SearchRequestOptionsRequest
 from .types.search_response import SearchResponse
 from .types.submit_completion_actual_request import SubmitCompletionActualRequest
 from .types.submit_workflow_execution_actual_request import SubmitWorkflowExecutionActualRequest
-from .core.client_wrapper import AsyncClientWrapper
-from .resources.ad_hoc.client import AsyncAdHocClient
-from .resources.container_images.client import AsyncContainerImagesClient
-from .resources.deployments.client import AsyncDeploymentsClient
-from .resources.document_indexes.client import AsyncDocumentIndexesClient
-from .resources.documents.client import AsyncDocumentsClient
-from .resources.folder_entities.client import AsyncFolderEntitiesClient
-from .resources.metric_definitions.client import AsyncMetricDefinitionsClient
-from .resources.ml_models.client import AsyncMlModelsClient
-from .resources.organizations.client import AsyncOrganizationsClient
-from .resources.prompts.client import AsyncPromptsClient
-from .resources.sandboxes.client import AsyncSandboxesClient
-from .resources.test_suite_runs.client import AsyncTestSuiteRunsClient
-from .resources.test_suites.client import AsyncTestSuitesClient
-from .resources.workflow_deployments.client import AsyncWorkflowDeploymentsClient
-from .resources.release_reviews.client import AsyncReleaseReviewsClient
-from .resources.workflow_sandboxes.client import AsyncWorkflowSandboxesClient
-from .resources.workflows.client import AsyncWorkflowsClient
-from .resources.workspace_secrets.client import AsyncWorkspaceSecretsClient
-from .resources.workspaces.client import AsyncWorkspacesClient
+from .types.vellum_variable_type import VellumVariableType
+from .types.workflow_execution_event_type import WorkflowExecutionEventType
+from .types.workflow_expand_meta_request import WorkflowExpandMetaRequest
+from .types.workflow_request_input_request import WorkflowRequestInputRequest
+from .types.workflow_stream_event import WorkflowStreamEvent
 
 # this is used as the default value for optional parameters
 OMIT = typing.cast(typing.Any, ...)
@@ -101,7 +84,11 @@ class Vellum:
 
 
 
+    api_version : typing.Optional[ApiVersionEnum]
     api_key : str
+    headers : typing.Optional[typing.Dict[str, str]]
+        Additional headers to send with every request.
+
     timeout : typing.Optional[float]
         The timeout to be used, in seconds, for requests. By default there is no timeout set, unless a custom httpx client is used, in which case this default is not enforced.
 
@@ -116,6 +103,7 @@ class Vellum:
     from vellum import Vellum
 
     client = Vellum(
+        api_version="YOUR_API_VERSION",
         api_key="YOUR_API_KEY",
     )
     """
@@ -124,15 +112,21 @@ class Vellum:
         self,
         *,
         environment: VellumEnvironment = VellumEnvironment.PRODUCTION,
+        api_version: typing.Optional[ApiVersionEnum] = os.getenv("VELLUM_API_VERSION", "2025-07-30"),
         api_key: str,
+        headers: typing.Optional[typing.Dict[str, str]] = None,
         timeout: typing.Optional[float] = None,
         follow_redirects: typing.Optional[bool] = True,
         httpx_client: typing.Optional[httpx.Client] = None,
     ):
-        _defaulted_timeout = timeout if timeout is not None else None if httpx_client is None else None
+        _defaulted_timeout = (
+            timeout if timeout is not None else None if httpx_client is None else httpx_client.timeout.read
+        )
         self._client_wrapper = SyncClientWrapper(
             environment=environment,
+            api_version=api_version,
             api_key=api_key,
+            headers=headers,
             httpx_client=httpx_client
             if httpx_client is not None
             else httpx.Client(timeout=_defaulted_timeout, follow_redirects=follow_redirects)
@@ -140,12 +134,18 @@ class Vellum:
             else httpx.Client(timeout=_defaulted_timeout),
             timeout=_defaulted_timeout,
         )
+        self._raw_client = RawVellum(client_wrapper=self._client_wrapper)
+        self.integrations = IntegrationsClient(client_wrapper=self._client_wrapper)
+        self.events = EventsClient(client_wrapper=self._client_wrapper)
         self.ad_hoc = AdHocClient(client_wrapper=self._client_wrapper)
         self.container_images = ContainerImagesClient(client_wrapper=self._client_wrapper)
         self.deployments = DeploymentsClient(client_wrapper=self._client_wrapper)
         self.document_indexes = DocumentIndexesClient(client_wrapper=self._client_wrapper)
         self.documents = DocumentsClient(client_wrapper=self._client_wrapper)
+        self.environments = EnvironmentsClient(client_wrapper=self._client_wrapper)
         self.folder_entities = FolderEntitiesClient(client_wrapper=self._client_wrapper)
+        self.integration_auth_configs = IntegrationAuthConfigsClient(client_wrapper=self._client_wrapper)
+        self.integration_providers = IntegrationProvidersClient(client_wrapper=self._client_wrapper)
         self.metric_definitions = MetricDefinitionsClient(client_wrapper=self._client_wrapper)
         self.ml_models = MlModelsClient(client_wrapper=self._client_wrapper)
         self.organizations = OrganizationsClient(client_wrapper=self._client_wrapper)
@@ -153,13 +153,25 @@ class Vellum:
         self.sandboxes = SandboxesClient(client_wrapper=self._client_wrapper)
         self.test_suite_runs = TestSuiteRunsClient(client_wrapper=self._client_wrapper)
         self.test_suites = TestSuitesClient(client_wrapper=self._client_wrapper)
+        self.uploaded_files = UploadedFilesClient(client_wrapper=self._client_wrapper)
         self.workflow_deployments = WorkflowDeploymentsClient(client_wrapper=self._client_wrapper)
-        self.release_reviews = ReleaseReviewsClient(client_wrapper=self._client_wrapper)
+        self.workflow_executions = WorkflowExecutionsClient(client_wrapper=self._client_wrapper)
         self.workflow_sandboxes = WorkflowSandboxesClient(client_wrapper=self._client_wrapper)
         self.workflows = WorkflowsClient(client_wrapper=self._client_wrapper)
         self.workspace_secrets = WorkspaceSecretsClient(client_wrapper=self._client_wrapper)
         self.workspaces = WorkspacesClient(client_wrapper=self._client_wrapper)
 
+    @property
+    def with_raw_response(self) -> RawVellum:
+        """
+        Retrieves a raw implementation of this client that returns raw responses.
+
+        Returns
+        -------
+        RawVellum
+        """
+        return self._raw_client
+
     def execute_api(
         self,
         *,
@@ -196,48 +208,22 @@ class Vellum:
         from vellum import Vellum
 
         client = Vellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
         client.execute_api(
-            url="url",
+            url="x",
         )
         """
-        _response = self._client_wrapper.httpx_client.request(
-            "v1/execute-api",
-            base_url=self._client_wrapper.get_environment().default,
-            method="POST",
-            json={
-                "url": url,
-                "method": method,
-                "body": convert_and_respect_annotation_metadata(
-                    object_=body, annotation=ExecuteApiRequestBody, direction="write"
-                ),
-                "headers": convert_and_respect_annotation_metadata(
-                    object_=headers, annotation=typing.Dict[str, ExecuteApiRequestHeadersValue], direction="write"
-                ),
-                "bearer_token": convert_and_respect_annotation_metadata(
-                    object_=bearer_token, annotation=ExecuteApiRequestBearerToken, direction="write"
-                ),
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = self._raw_client.execute_api(
+            url=url,
+            method=method,
+            body=body,
+            headers=headers,
+            bearer_token=bearer_token,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    ExecuteApiResponse,
-                    parse_obj_as(
-                        type_=ExecuteApiResponse,  # type: ignore
-                        object_=_response.json(),
-                    ),
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     def execute_code(
         self,
@@ -275,70 +261,44 @@ class Vellum:
         from vellum import CodeExecutionPackage, StringInput, Vellum
 
         client = Vellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
         client.execute_code(
-            code="code",
+            code="x",
             runtime="PYTHON_3_11_6",
             input_values=[
                 StringInput(
-                    name="name",
+                    name="x",
                     value="value",
-                )
+                ),
+                StringInput(
+                    name="x",
+                    value="value",
+                ),
             ],
             packages=[
                 CodeExecutionPackage(
                     version="version",
                     name="name",
-                )
+                ),
+                CodeExecutionPackage(
+                    version="version",
+                    name="name",
+                ),
             ],
             output_type="STRING",
         )
         """
-        _response = self._client_wrapper.httpx_client.request(
-            "v1/execute-code",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "code": code,
-                "runtime": runtime,
-                "input_values": convert_and_respect_annotation_metadata(
-                    object_=input_values, annotation=typing.Sequence[CodeExecutorInput], direction="write"
-                ),
-                "packages": convert_and_respect_annotation_metadata(
-                    object_=packages, annotation=typing.Sequence[CodeExecutionPackage], direction="write"
-                ),
-                "output_type": output_type,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = self._raw_client.execute_code(
+            code=code,
+            runtime=runtime,
+            input_values=input_values,
+            packages=packages,
+            output_type=output_type,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    CodeExecutorResponse,
-                    parse_obj_as(
-                        type_=CodeExecutorResponse,  # type: ignore
-                        object_=_response.json(),
-                    ),
-                )
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     def execute_prompt(
         self,
@@ -399,97 +359,35 @@ class Vellum:
         from vellum import StringInputRequest, Vellum
 
         client = Vellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
         client.execute_prompt(
             inputs=[
                 StringInputRequest(
-                    name="name",
+                    name="x",
+                    value="value",
+                ),
+                StringInputRequest(
+                    name="x",
                     value="value",
-                )
+                ),
             ],
         )
         """
-        _response = self._client_wrapper.httpx_client.request(
-            "v1/execute-prompt",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "inputs": convert_and_respect_annotation_metadata(
-                    object_=inputs, annotation=typing.Sequence[PromptDeploymentInputRequest], direction="write"
-                ),
-                "prompt_deployment_id": prompt_deployment_id,
-                "prompt_deployment_name": prompt_deployment_name,
-                "release_tag": release_tag,
-                "external_id": external_id,
-                "expand_meta": convert_and_respect_annotation_metadata(
-                    object_=expand_meta, annotation=PromptDeploymentExpandMetaRequest, direction="write"
-                ),
-                "raw_overrides": convert_and_respect_annotation_metadata(
-                    object_=raw_overrides, annotation=RawPromptExecutionOverridesRequest, direction="write"
-                ),
-                "expand_raw": expand_raw,
-                "metadata": metadata,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = self._raw_client.execute_prompt(
+            inputs=inputs,
+            prompt_deployment_id=prompt_deployment_id,
+            prompt_deployment_name=prompt_deployment_name,
+            release_tag=release_tag,
+            external_id=external_id,
+            expand_meta=expand_meta,
+            raw_overrides=raw_overrides,
+            expand_raw=expand_raw,
+            metadata=metadata,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    ExecutePromptResponse,
-                    parse_obj_as(
-                        type_=ExecutePromptResponse,  # type: ignore
-                        object_=_response.json(),
-                    ),
-                )
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 403:
-                raise ForbiddenError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 404:
-                raise NotFoundError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 500:
-                raise InternalServerError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     def execute_prompt_stream(
         self,
@@ -550,107 +448,37 @@ class Vellum:
         from vellum import StringInputRequest, Vellum
 
         client = Vellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
         response = client.execute_prompt_stream(
             inputs=[
                 StringInputRequest(
-                    name="name",
+                    name="x",
                     value="value",
-                )
+                ),
+                StringInputRequest(
+                    name="x",
+                    value="value",
+                ),
             ],
         )
         for chunk in response:
             yield chunk
         """
-        with self._client_wrapper.httpx_client.stream(
-            "v1/execute-prompt-stream",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "inputs": convert_and_respect_annotation_metadata(
-                    object_=inputs, annotation=typing.Sequence[PromptDeploymentInputRequest], direction="write"
-                ),
-                "prompt_deployment_id": prompt_deployment_id,
-                "prompt_deployment_name": prompt_deployment_name,
-                "release_tag": release_tag,
-                "external_id": external_id,
-                "expand_meta": convert_and_respect_annotation_metadata(
-                    object_=expand_meta, annotation=PromptDeploymentExpandMetaRequest, direction="write"
-                ),
-                "raw_overrides": convert_and_respect_annotation_metadata(
-                    object_=raw_overrides, annotation=RawPromptExecutionOverridesRequest, direction="write"
-                ),
-                "expand_raw": expand_raw,
-                "metadata": metadata,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        with self._raw_client.execute_prompt_stream(
+            inputs=inputs,
+            prompt_deployment_id=prompt_deployment_id,
+            prompt_deployment_name=prompt_deployment_name,
+            release_tag=release_tag,
+            external_id=external_id,
+            expand_meta=expand_meta,
+            raw_overrides=raw_overrides,
+            expand_raw=expand_raw,
+            metadata=metadata,
             request_options=request_options,
-            omit=OMIT,
-        ) as _response:
-            try:
-                if 200 <= _response.status_code < 300:
-                    for _text in _response.iter_lines():
-                        try:
-                            if len(_text) == 0:
-                                continue
-                            yield typing.cast(
-                                ExecutePromptEvent,
-                                parse_obj_as(
-                                    type_=ExecutePromptEvent,  # type: ignore
-                                    object_=json.loads(_text),
-                                ),
-                            )
-                        except Exception:
-                            pass
-                    return
-                _response.read()
-                if _response.status_code == 400:
-                    raise BadRequestError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 403:
-                    raise ForbiddenError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 404:
-                    raise NotFoundError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 500:
-                    raise InternalServerError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                _response_json = _response.json()
-            except JSONDecodeError:
-                raise ApiError(status_code=_response.status_code, body=_response.text)
-            raise ApiError(status_code=_response.status_code, body=_response_json)
+        ) as r:
+            yield from r.data
 
     def execute_workflow(
         self,
@@ -662,6 +490,7 @@ class Vellum:
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
         metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
+        previous_execution_id: typing.Optional[str] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> ExecuteWorkflowResponse:
         """
@@ -690,6 +519,9 @@ class Vellum:
         metadata : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
             Arbitrary JSON metadata associated with this request. Can be used to capture additional monitoring data such as user id, session id, etc. for future analysis.
 
+        previous_execution_id : typing.Optional[str]
+            The ID of a previous Workflow Execution to reference for initial State loading.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -703,83 +535,113 @@ class Vellum:
         from vellum import Vellum, WorkflowRequestStringInputRequest
 
         client = Vellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
         client.execute_workflow(
             inputs=[
                 WorkflowRequestStringInputRequest(
-                    name="name",
+                    name="x",
+                    value="value",
+                ),
+                WorkflowRequestStringInputRequest(
+                    name="x",
                     value="value",
-                )
+                ),
             ],
         )
         """
-        _response = self._client_wrapper.httpx_client.request(
-            "v1/execute-workflow",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "inputs": convert_and_respect_annotation_metadata(
-                    object_=inputs, annotation=typing.Sequence[WorkflowRequestInputRequest], direction="write"
+        _response = self._raw_client.execute_workflow(
+            inputs=inputs,
+            expand_meta=expand_meta,
+            workflow_deployment_id=workflow_deployment_id,
+            workflow_deployment_name=workflow_deployment_name,
+            release_tag=release_tag,
+            external_id=external_id,
+            metadata=metadata,
+            previous_execution_id=previous_execution_id,
+            request_options=request_options,
+        )
+        return _response.data
+
+    def execute_workflow_async(
+        self,
+        *,
+        inputs: typing.Sequence[WorkflowRequestInputRequest],
+        workflow_deployment_id: typing.Optional[str] = OMIT,
+        workflow_deployment_name: typing.Optional[str] = OMIT,
+        release_tag: typing.Optional[str] = OMIT,
+        external_id: typing.Optional[str] = OMIT,
+        previous_execution_id: typing.Optional[str] = OMIT,
+        metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> ExecuteWorkflowAsyncResponse:
+        """
+        Executes a deployed Workflow asynchronously and returns the execution ID.
+
+        Parameters
+        ----------
+        inputs : typing.Sequence[WorkflowRequestInputRequest]
+            The list of inputs defined in the Workflow's Deployment with their corresponding values.
+
+        workflow_deployment_id : typing.Optional[str]
+            The ID of the Workflow Deployment. Must provide either this or workflow_deployment_name.
+
+        workflow_deployment_name : typing.Optional[str]
+            The name of the Workflow Deployment. Must provide either this or workflow_deployment_id.
+
+        release_tag : typing.Optional[str]
+            Optionally specify a release tag if you want to pin to a specific release of the Workflow Deployment
+
+        external_id : typing.Optional[str]
+            Optionally include a unique identifier for tracking purposes. Must be unique within a given Workspace.
+
+        previous_execution_id : typing.Optional[str]
+            The ID of a previous Workflow Execution to reference for initial State loading.
+
+        metadata : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
+            Arbitrary JSON metadata associated with this request. Can be used to capture additional monitoring data such as user id, session id, etc. for future analysis.
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Returns
+        -------
+        ExecuteWorkflowAsyncResponse
+
+
+        Examples
+        --------
+        from vellum import Vellum, WorkflowRequestStringInputRequest
+
+        client = Vellum(
+            api_version="YOUR_API_VERSION",
+            api_key="YOUR_API_KEY",
+        )
+        client.execute_workflow_async(
+            inputs=[
+                WorkflowRequestStringInputRequest(
+                    name="x",
+                    value="value",
                 ),
-                "expand_meta": convert_and_respect_annotation_metadata(
-                    object_=expand_meta, annotation=WorkflowExpandMetaRequest, direction="write"
+                WorkflowRequestStringInputRequest(
+                    name="x",
+                    value="value",
                 ),
-                "workflow_deployment_id": workflow_deployment_id,
-                "workflow_deployment_name": workflow_deployment_name,
-                "release_tag": release_tag,
-                "external_id": external_id,
-                "metadata": metadata,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+            ],
+        )
+        """
+        _response = self._raw_client.execute_workflow_async(
+            inputs=inputs,
+            workflow_deployment_id=workflow_deployment_id,
+            workflow_deployment_name=workflow_deployment_name,
+            release_tag=release_tag,
+            external_id=external_id,
+            previous_execution_id=previous_execution_id,
+            metadata=metadata,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    ExecuteWorkflowResponse,
-                    parse_obj_as(
-                        type_=ExecuteWorkflowResponse,  # type: ignore
-                        object_=_response.json(),
-                    ),
-                )
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 404:
-                raise NotFoundError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 500:
-                raise InternalServerError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     def execute_workflow_stream(
         self,
@@ -792,6 +654,7 @@ class Vellum:
         external_id: typing.Optional[str] = OMIT,
         event_types: typing.Optional[typing.Sequence[WorkflowExecutionEventType]] = OMIT,
         metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
+        previous_execution_id: typing.Optional[str] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[WorkflowStreamEvent]:
         """
@@ -823,6 +686,9 @@ class Vellum:
         metadata : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
             Arbitrary JSON metadata associated with this request. Can be used to capture additional monitoring data such as user id, session id, etc. for future analysis.
 
+        previous_execution_id : typing.Optional[str]
+            The ID of a previous Workflow Execution to reference for initial State loading.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -836,94 +702,37 @@ class Vellum:
         from vellum import Vellum, WorkflowRequestStringInputRequest
 
         client = Vellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
         response = client.execute_workflow_stream(
             inputs=[
                 WorkflowRequestStringInputRequest(
-                    name="name",
+                    name="x",
                     value="value",
-                )
+                ),
+                WorkflowRequestStringInputRequest(
+                    name="x",
+                    value="value",
+                ),
             ],
         )
         for chunk in response:
             yield chunk
         """
-        with self._client_wrapper.httpx_client.stream(
-            "v1/execute-workflow-stream",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "inputs": convert_and_respect_annotation_metadata(
-                    object_=inputs, annotation=typing.Sequence[WorkflowRequestInputRequest], direction="write"
-                ),
-                "expand_meta": convert_and_respect_annotation_metadata(
-                    object_=expand_meta, annotation=WorkflowExpandMetaRequest, direction="write"
-                ),
-                "workflow_deployment_id": workflow_deployment_id,
-                "workflow_deployment_name": workflow_deployment_name,
-                "release_tag": release_tag,
-                "external_id": external_id,
-                "event_types": event_types,
-                "metadata": metadata,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        with self._raw_client.execute_workflow_stream(
+            inputs=inputs,
+            expand_meta=expand_meta,
+            workflow_deployment_id=workflow_deployment_id,
+            workflow_deployment_name=workflow_deployment_name,
+            release_tag=release_tag,
+            external_id=external_id,
+            event_types=event_types,
+            metadata=metadata,
+            previous_execution_id=previous_execution_id,
             request_options=request_options,
-            omit=OMIT,
-        ) as _response:
-            try:
-                if 200 <= _response.status_code < 300:
-                    for _text in _response.iter_lines():
-                        try:
-                            if len(_text) == 0:
-                                continue
-                            yield typing.cast(
-                                WorkflowStreamEvent,
-                                parse_obj_as(
-                                    type_=WorkflowStreamEvent,  # type: ignore
-                                    object_=json.loads(_text),
-                                ),
-                            )
-                        except Exception:
-                            pass
-                    return
-                _response.read()
-                if _response.status_code == 400:
-                    raise BadRequestError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 404:
-                    raise NotFoundError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 500:
-                    raise InternalServerError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                _response_json = _response.json()
-            except JSONDecodeError:
-                raise ApiError(status_code=_response.status_code, body=_response.text)
-            raise ApiError(status_code=_response.status_code, body=_response_json)
+        ) as r:
+            yield from r.data
 
     def generate(
         self,
@@ -967,89 +776,28 @@ class Vellum:
         from vellum import GenerateRequest, Vellum
 
         client = Vellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
         client.generate(
             requests=[
                 GenerateRequest(
-                    input_values={"key": "value"},
-                )
+                    input_values={"input_values": {"key": "value"}},
+                ),
+                GenerateRequest(
+                    input_values={"input_values": {"key": "value"}},
+                ),
             ],
         )
         """
-        _response = self._client_wrapper.httpx_client.request(
-            "v1/generate",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "deployment_id": deployment_id,
-                "deployment_name": deployment_name,
-                "requests": convert_and_respect_annotation_metadata(
-                    object_=requests, annotation=typing.Sequence[GenerateRequest], direction="write"
-                ),
-                "options": convert_and_respect_annotation_metadata(
-                    object_=options, annotation=GenerateOptionsRequest, direction="write"
-                ),
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = self._raw_client.generate(
+            requests=requests,
+            deployment_id=deployment_id,
+            deployment_name=deployment_name,
+            options=options,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    GenerateResponse,
-                    parse_obj_as(
-                        type_=GenerateResponse,  # type: ignore
-                        object_=_response.json(),
-                    ),
-                )
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 403:
-                raise ForbiddenError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 404:
-                raise NotFoundError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 500:
-                raise InternalServerError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     def generate_stream(
         self,
@@ -1093,99 +841,30 @@ class Vellum:
         from vellum import GenerateRequest, Vellum
 
         client = Vellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
         response = client.generate_stream(
             requests=[
                 GenerateRequest(
-                    input_values={"key": "value"},
-                )
+                    input_values={"input_values": {"key": "value"}},
+                ),
+                GenerateRequest(
+                    input_values={"input_values": {"key": "value"}},
+                ),
             ],
         )
         for chunk in response:
             yield chunk
         """
-        with self._client_wrapper.httpx_client.stream(
-            "v1/generate-stream",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "deployment_id": deployment_id,
-                "deployment_name": deployment_name,
-                "requests": convert_and_respect_annotation_metadata(
-                    object_=requests, annotation=typing.Sequence[GenerateRequest], direction="write"
-                ),
-                "options": convert_and_respect_annotation_metadata(
-                    object_=options, annotation=GenerateOptionsRequest, direction="write"
-                ),
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        with self._raw_client.generate_stream(
+            requests=requests,
+            deployment_id=deployment_id,
+            deployment_name=deployment_name,
+            options=options,
             request_options=request_options,
-            omit=OMIT,
-        ) as _response:
-            try:
-                if 200 <= _response.status_code < 300:
-                    for _text in _response.iter_lines():
-                        try:
-                            if len(_text) == 0:
-                                continue
-                            yield typing.cast(
-                                GenerateStreamResponse,
-                                parse_obj_as(
-                                    type_=GenerateStreamResponse,  # type: ignore
-                                    object_=json.loads(_text),
-                                ),
-                            )
-                        except Exception:
-                            pass
-                    return
-                _response.read()
-                if _response.status_code == 400:
-                    raise BadRequestError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 403:
-                    raise ForbiddenError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 404:
-                    raise NotFoundError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 500:
-                    raise InternalServerError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                _response_json = _response.json()
-            except JSONDecodeError:
-                raise ApiError(status_code=_response.status_code, body=_response.text)
-            raise ApiError(status_code=_response.status_code, body=_response_json)
+        ) as r:
+            yield from r.data
 
     def search(
         self,
@@ -1230,74 +909,22 @@ class Vellum:
         from vellum import Vellum
 
         client = Vellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
         client.search(
-            query="query",
+            query="x",
         )
         """
-        _response = self._client_wrapper.httpx_client.request(
-            "v1/search",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "index_id": index_id,
-                "index_name": index_name,
-                "query": query,
-                "options": convert_and_respect_annotation_metadata(
-                    object_=options, annotation=SearchRequestOptionsRequest, direction="write"
-                ),
-                "document_index": document_index,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = self._raw_client.search(
+            query=query,
+            index_id=index_id,
+            index_name=index_name,
+            options=options,
+            document_index=document_index,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    SearchResponse,
-                    parse_obj_as(
-                        type_=SearchResponse,  # type: ignore
-                        object_=_response.json(),
-                    ),
-                )
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 404:
-                raise NotFoundError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 500:
-                raise InternalServerError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     def submit_completion_actuals(
         self,
@@ -1333,66 +960,20 @@ class Vellum:
         from vellum import SubmitCompletionActualRequest, Vellum
 
         client = Vellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
         client.submit_completion_actuals(
-            actuals=[SubmitCompletionActualRequest()],
+            actuals=[SubmitCompletionActualRequest(), SubmitCompletionActualRequest()],
         )
         """
-        _response = self._client_wrapper.httpx_client.request(
-            "v1/submit-completion-actuals",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "deployment_id": deployment_id,
-                "deployment_name": deployment_name,
-                "actuals": convert_and_respect_annotation_metadata(
-                    object_=actuals, annotation=typing.Sequence[SubmitCompletionActualRequest], direction="write"
-                ),
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = self._raw_client.submit_completion_actuals(
+            actuals=actuals,
+            deployment_id=deployment_id,
+            deployment_name=deployment_name,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 404:
-                raise NotFoundError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 500:
-                raise InternalServerError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     def submit_workflow_execution_actuals(
         self,
@@ -1430,36 +1011,20 @@ class Vellum:
         from vellum import Vellum, WorkflowExecutionActualStringRequest
 
         client = Vellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
         client.submit_workflow_execution_actuals(
-            actuals=[WorkflowExecutionActualStringRequest()],
+            actuals=[
+                WorkflowExecutionActualStringRequest(),
+                WorkflowExecutionActualStringRequest(),
+            ],
         )
         """
-        _response = self._client_wrapper.httpx_client.request(
-            "v1/submit-workflow-execution-actuals",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "actuals": convert_and_respect_annotation_metadata(
-                    object_=actuals, annotation=typing.Sequence[SubmitWorkflowExecutionActualRequest], direction="write"
-                ),
-                "execution_id": execution_id,
-                "external_id": external_id,
-            },
-            headers={
-                "content-type": "application/json",
-            },
-            request_options=request_options,
-            omit=OMIT,
+        _response = self._raw_client.submit_workflow_execution_actuals(
+            actuals=actuals, execution_id=execution_id, external_id=external_id, request_options=request_options
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
 
 class AsyncVellum:
@@ -1477,7 +1042,11 @@ class AsyncVellum:
 
 
 
+    api_version : typing.Optional[ApiVersionEnum]
     api_key : str
+    headers : typing.Optional[typing.Dict[str, str]]
+        Additional headers to send with every request.
+
     timeout : typing.Optional[float]
         The timeout to be used, in seconds, for requests. By default there is no timeout set, unless a custom httpx client is used, in which case this default is not enforced.
 
@@ -1492,6 +1061,7 @@ class AsyncVellum:
     from vellum import AsyncVellum
 
     client = AsyncVellum(
+        api_version="YOUR_API_VERSION",
         api_key="YOUR_API_KEY",
     )
     """
@@ -1500,15 +1070,21 @@ class AsyncVellum:
         self,
         *,
         environment: VellumEnvironment = VellumEnvironment.PRODUCTION,
+        api_version: typing.Optional[ApiVersionEnum] = os.getenv("VELLUM_API_VERSION", "2025-07-30"),
         api_key: str,
+        headers: typing.Optional[typing.Dict[str, str]] = None,
         timeout: typing.Optional[float] = None,
         follow_redirects: typing.Optional[bool] = True,
         httpx_client: typing.Optional[httpx.AsyncClient] = None,
     ):
-        _defaulted_timeout = timeout if timeout is not None else None if httpx_client is None else None
+        _defaulted_timeout = (
+            timeout if timeout is not None else None if httpx_client is None else httpx_client.timeout.read
+        )
         self._client_wrapper = AsyncClientWrapper(
             environment=environment,
+            api_version=api_version,
             api_key=api_key,
+            headers=headers,
             httpx_client=httpx_client
             if httpx_client is not None
             else httpx.AsyncClient(timeout=_defaulted_timeout, follow_redirects=follow_redirects)
@@ -1516,12 +1092,18 @@ class AsyncVellum:
             else httpx.AsyncClient(timeout=_defaulted_timeout),
             timeout=_defaulted_timeout,
         )
+        self._raw_client = AsyncRawVellum(client_wrapper=self._client_wrapper)
+        self.integrations = AsyncIntegrationsClient(client_wrapper=self._client_wrapper)
+        self.events = AsyncEventsClient(client_wrapper=self._client_wrapper)
         self.ad_hoc = AsyncAdHocClient(client_wrapper=self._client_wrapper)
         self.container_images = AsyncContainerImagesClient(client_wrapper=self._client_wrapper)
         self.deployments = AsyncDeploymentsClient(client_wrapper=self._client_wrapper)
         self.document_indexes = AsyncDocumentIndexesClient(client_wrapper=self._client_wrapper)
         self.documents = AsyncDocumentsClient(client_wrapper=self._client_wrapper)
+        self.environments = AsyncEnvironmentsClient(client_wrapper=self._client_wrapper)
         self.folder_entities = AsyncFolderEntitiesClient(client_wrapper=self._client_wrapper)
+        self.integration_auth_configs = AsyncIntegrationAuthConfigsClient(client_wrapper=self._client_wrapper)
+        self.integration_providers = AsyncIntegrationProvidersClient(client_wrapper=self._client_wrapper)
         self.metric_definitions = AsyncMetricDefinitionsClient(client_wrapper=self._client_wrapper)
         self.ml_models = AsyncMlModelsClient(client_wrapper=self._client_wrapper)
         self.organizations = AsyncOrganizationsClient(client_wrapper=self._client_wrapper)
@@ -1529,13 +1111,25 @@ class AsyncVellum:
         self.sandboxes = AsyncSandboxesClient(client_wrapper=self._client_wrapper)
         self.test_suite_runs = AsyncTestSuiteRunsClient(client_wrapper=self._client_wrapper)
         self.test_suites = AsyncTestSuitesClient(client_wrapper=self._client_wrapper)
+        self.uploaded_files = AsyncUploadedFilesClient(client_wrapper=self._client_wrapper)
         self.workflow_deployments = AsyncWorkflowDeploymentsClient(client_wrapper=self._client_wrapper)
-        self.release_reviews = AsyncReleaseReviewsClient(client_wrapper=self._client_wrapper)
+        self.workflow_executions = AsyncWorkflowExecutionsClient(client_wrapper=self._client_wrapper)
         self.workflow_sandboxes = AsyncWorkflowSandboxesClient(client_wrapper=self._client_wrapper)
         self.workflows = AsyncWorkflowsClient(client_wrapper=self._client_wrapper)
         self.workspace_secrets = AsyncWorkspaceSecretsClient(client_wrapper=self._client_wrapper)
         self.workspaces = AsyncWorkspacesClient(client_wrapper=self._client_wrapper)
 
+    @property
+    def with_raw_response(self) -> AsyncRawVellum:
+        """
+        Retrieves a raw implementation of this client that returns raw responses.
+
+        Returns
+        -------
+        AsyncRawVellum
+        """
+        return self._raw_client
+
     async def execute_api(
         self,
         *,
@@ -1574,54 +1168,28 @@ class AsyncVellum:
         from vellum import AsyncVellum
 
         client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
 
 
         async def main() -> None:
             await client.execute_api(
-                url="url",
+                url="x",
             )
 
 
         asyncio.run(main())
         """
-        _response = await self._client_wrapper.httpx_client.request(
-            "v1/execute-api",
-            base_url=self._client_wrapper.get_environment().default,
-            method="POST",
-            json={
-                "url": url,
-                "method": method,
-                "body": convert_and_respect_annotation_metadata(
-                    object_=body, annotation=ExecuteApiRequestBody, direction="write"
-                ),
-                "headers": convert_and_respect_annotation_metadata(
-                    object_=headers, annotation=typing.Dict[str, ExecuteApiRequestHeadersValue], direction="write"
-                ),
-                "bearer_token": convert_and_respect_annotation_metadata(
-                    object_=bearer_token, annotation=ExecuteApiRequestBearerToken, direction="write"
-                ),
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = await self._raw_client.execute_api(
+            url=url,
+            method=method,
+            body=body,
+            headers=headers,
+            bearer_token=bearer_token,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    ExecuteApiResponse,
-                    parse_obj_as(
-                        type_=ExecuteApiResponse,  # type: ignore
-                        object_=_response.json(),
-                    ),
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     async def execute_code(
         self,
@@ -1661,25 +1229,34 @@ class AsyncVellum:
         from vellum import AsyncVellum, CodeExecutionPackage, StringInput
 
         client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
 
 
         async def main() -> None:
             await client.execute_code(
-                code="code",
+                code="x",
                 runtime="PYTHON_3_11_6",
                 input_values=[
                     StringInput(
-                        name="name",
+                        name="x",
                         value="value",
-                    )
+                    ),
+                    StringInput(
+                        name="x",
+                        value="value",
+                    ),
                 ],
                 packages=[
                     CodeExecutionPackage(
                         version="version",
                         name="name",
-                    )
+                    ),
+                    CodeExecutionPackage(
+                        version="version",
+                        name="name",
+                    ),
                 ],
                 output_type="STRING",
             )
@@ -1687,50 +1264,15 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        _response = await self._client_wrapper.httpx_client.request(
-            "v1/execute-code",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "code": code,
-                "runtime": runtime,
-                "input_values": convert_and_respect_annotation_metadata(
-                    object_=input_values, annotation=typing.Sequence[CodeExecutorInput], direction="write"
-                ),
-                "packages": convert_and_respect_annotation_metadata(
-                    object_=packages, annotation=typing.Sequence[CodeExecutionPackage], direction="write"
-                ),
-                "output_type": output_type,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = await self._raw_client.execute_code(
+            code=code,
+            runtime=runtime,
+            input_values=input_values,
+            packages=packages,
+            output_type=output_type,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    CodeExecutorResponse,
-                    parse_obj_as(
-                        type_=CodeExecutorResponse,  # type: ignore
-                        object_=_response.json(),
-                    ),
-                )
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     async def execute_prompt(
         self,
@@ -1793,6 +1335,7 @@ class AsyncVellum:
         from vellum import AsyncVellum, StringInputRequest
 
         client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
 
@@ -1801,95 +1344,32 @@ class AsyncVellum:
             await client.execute_prompt(
                 inputs=[
                     StringInputRequest(
-                        name="name",
+                        name="x",
                         value="value",
-                    )
+                    ),
+                    StringInputRequest(
+                        name="x",
+                        value="value",
+                    ),
                 ],
             )
 
 
         asyncio.run(main())
         """
-        _response = await self._client_wrapper.httpx_client.request(
-            "v1/execute-prompt",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "inputs": convert_and_respect_annotation_metadata(
-                    object_=inputs, annotation=typing.Sequence[PromptDeploymentInputRequest], direction="write"
-                ),
-                "prompt_deployment_id": prompt_deployment_id,
-                "prompt_deployment_name": prompt_deployment_name,
-                "release_tag": release_tag,
-                "external_id": external_id,
-                "expand_meta": convert_and_respect_annotation_metadata(
-                    object_=expand_meta, annotation=PromptDeploymentExpandMetaRequest, direction="write"
-                ),
-                "raw_overrides": convert_and_respect_annotation_metadata(
-                    object_=raw_overrides, annotation=RawPromptExecutionOverridesRequest, direction="write"
-                ),
-                "expand_raw": expand_raw,
-                "metadata": metadata,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = await self._raw_client.execute_prompt(
+            inputs=inputs,
+            prompt_deployment_id=prompt_deployment_id,
+            prompt_deployment_name=prompt_deployment_name,
+            release_tag=release_tag,
+            external_id=external_id,
+            expand_meta=expand_meta,
+            raw_overrides=raw_overrides,
+            expand_raw=expand_raw,
+            metadata=metadata,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    ExecutePromptResponse,
-                    parse_obj_as(
-                        type_=ExecutePromptResponse,  # type: ignore
-                        object_=_response.json(),
-                    ),
-                )
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 403:
-                raise ForbiddenError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 404:
-                raise NotFoundError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 500:
-                raise InternalServerError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     async def execute_prompt_stream(
         self,
@@ -1952,6 +1432,7 @@ class AsyncVellum:
         from vellum import AsyncVellum, StringInputRequest
 
         client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
 
@@ -1960,9 +1441,13 @@ class AsyncVellum:
             response = await client.execute_prompt_stream(
                 inputs=[
                     StringInputRequest(
-                        name="name",
+                        name="x",
+                        value="value",
+                    ),
+                    StringInputRequest(
+                        name="x",
                         value="value",
-                    )
+                    ),
                 ],
             )
             async for chunk in response:
@@ -1971,94 +1456,20 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        async with self._client_wrapper.httpx_client.stream(
-            "v1/execute-prompt-stream",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "inputs": convert_and_respect_annotation_metadata(
-                    object_=inputs, annotation=typing.Sequence[PromptDeploymentInputRequest], direction="write"
-                ),
-                "prompt_deployment_id": prompt_deployment_id,
-                "prompt_deployment_name": prompt_deployment_name,
-                "release_tag": release_tag,
-                "external_id": external_id,
-                "expand_meta": convert_and_respect_annotation_metadata(
-                    object_=expand_meta, annotation=PromptDeploymentExpandMetaRequest, direction="write"
-                ),
-                "raw_overrides": convert_and_respect_annotation_metadata(
-                    object_=raw_overrides, annotation=RawPromptExecutionOverridesRequest, direction="write"
-                ),
-                "expand_raw": expand_raw,
-                "metadata": metadata,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        async with self._raw_client.execute_prompt_stream(
+            inputs=inputs,
+            prompt_deployment_id=prompt_deployment_id,
+            prompt_deployment_name=prompt_deployment_name,
+            release_tag=release_tag,
+            external_id=external_id,
+            expand_meta=expand_meta,
+            raw_overrides=raw_overrides,
+            expand_raw=expand_raw,
+            metadata=metadata,
             request_options=request_options,
-            omit=OMIT,
-        ) as _response:
-            try:
-                if 200 <= _response.status_code < 300:
-                    async for _text in _response.aiter_lines():
-                        try:
-                            if len(_text) == 0:
-                                continue
-                            yield typing.cast(
-                                ExecutePromptEvent,
-                                parse_obj_as(
-                                    type_=ExecutePromptEvent,  # type: ignore
-                                    object_=json.loads(_text),
-                                ),
-                            )
-                        except Exception:
-                            pass
-                    return
-                await _response.aread()
-                if _response.status_code == 400:
-                    raise BadRequestError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 403:
-                    raise ForbiddenError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 404:
-                    raise NotFoundError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 500:
-                    raise InternalServerError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                _response_json = _response.json()
-            except JSONDecodeError:
-                raise ApiError(status_code=_response.status_code, body=_response.text)
-            raise ApiError(status_code=_response.status_code, body=_response_json)
+        ) as r:
+            async for _chunk in r.data:
+                yield _chunk
 
     async def execute_workflow(
         self,
@@ -2070,6 +1481,7 @@ class AsyncVellum:
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
         metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
+        previous_execution_id: typing.Optional[str] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> ExecuteWorkflowResponse:
         """
@@ -2098,6 +1510,9 @@ class AsyncVellum:
         metadata : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
             Arbitrary JSON metadata associated with this request. Can be used to capture additional monitoring data such as user id, session id, etc. for future analysis.
 
+        previous_execution_id : typing.Optional[str]
+            The ID of a previous Workflow Execution to reference for initial State loading.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -2113,6 +1528,7 @@ class AsyncVellum:
         from vellum import AsyncVellum, WorkflowRequestStringInputRequest
 
         client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
 
@@ -2121,81 +1537,118 @@ class AsyncVellum:
             await client.execute_workflow(
                 inputs=[
                     WorkflowRequestStringInputRequest(
-                        name="name",
+                        name="x",
                         value="value",
-                    )
+                    ),
+                    WorkflowRequestStringInputRequest(
+                        name="x",
+                        value="value",
+                    ),
                 ],
             )
 
 
         asyncio.run(main())
         """
-        _response = await self._client_wrapper.httpx_client.request(
-            "v1/execute-workflow",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "inputs": convert_and_respect_annotation_metadata(
-                    object_=inputs, annotation=typing.Sequence[WorkflowRequestInputRequest], direction="write"
-                ),
-                "expand_meta": convert_and_respect_annotation_metadata(
-                    object_=expand_meta, annotation=WorkflowExpandMetaRequest, direction="write"
-                ),
-                "workflow_deployment_id": workflow_deployment_id,
-                "workflow_deployment_name": workflow_deployment_name,
-                "release_tag": release_tag,
-                "external_id": external_id,
-                "metadata": metadata,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = await self._raw_client.execute_workflow(
+            inputs=inputs,
+            expand_meta=expand_meta,
+            workflow_deployment_id=workflow_deployment_id,
+            workflow_deployment_name=workflow_deployment_name,
+            release_tag=release_tag,
+            external_id=external_id,
+            metadata=metadata,
+            previous_execution_id=previous_execution_id,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    ExecuteWorkflowResponse,
-                    parse_obj_as(
-                        type_=ExecuteWorkflowResponse,  # type: ignore
-                        object_=_response.json(),
+        return _response.data
+
+    async def execute_workflow_async(
+        self,
+        *,
+        inputs: typing.Sequence[WorkflowRequestInputRequest],
+        workflow_deployment_id: typing.Optional[str] = OMIT,
+        workflow_deployment_name: typing.Optional[str] = OMIT,
+        release_tag: typing.Optional[str] = OMIT,
+        external_id: typing.Optional[str] = OMIT,
+        previous_execution_id: typing.Optional[str] = OMIT,
+        metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> ExecuteWorkflowAsyncResponse:
+        """
+        Executes a deployed Workflow asynchronously and returns the execution ID.
+
+        Parameters
+        ----------
+        inputs : typing.Sequence[WorkflowRequestInputRequest]
+            The list of inputs defined in the Workflow's Deployment with their corresponding values.
+
+        workflow_deployment_id : typing.Optional[str]
+            The ID of the Workflow Deployment. Must provide either this or workflow_deployment_name.
+
+        workflow_deployment_name : typing.Optional[str]
+            The name of the Workflow Deployment. Must provide either this or workflow_deployment_id.
+
+        release_tag : typing.Optional[str]
+            Optionally specify a release tag if you want to pin to a specific release of the Workflow Deployment
+
+        external_id : typing.Optional[str]
+            Optionally include a unique identifier for tracking purposes. Must be unique within a given Workspace.
+
+        previous_execution_id : typing.Optional[str]
+            The ID of a previous Workflow Execution to reference for initial State loading.
+
+        metadata : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
+            Arbitrary JSON metadata associated with this request. Can be used to capture additional monitoring data such as user id, session id, etc. for future analysis.
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Returns
+        -------
+        ExecuteWorkflowAsyncResponse
+
+
+        Examples
+        --------
+        import asyncio
+
+        from vellum import AsyncVellum, WorkflowRequestStringInputRequest
+
+        client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
+            api_key="YOUR_API_KEY",
+        )
+
+
+        async def main() -> None:
+            await client.execute_workflow_async(
+                inputs=[
+                    WorkflowRequestStringInputRequest(
+                        name="x",
+                        value="value",
                     ),
-                )
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 404:
-                raise NotFoundError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 500:
-                raise InternalServerError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+                    WorkflowRequestStringInputRequest(
+                        name="x",
+                        value="value",
+                    ),
+                ],
+            )
+
+
+        asyncio.run(main())
+        """
+        _response = await self._raw_client.execute_workflow_async(
+            inputs=inputs,
+            workflow_deployment_id=workflow_deployment_id,
+            workflow_deployment_name=workflow_deployment_name,
+            release_tag=release_tag,
+            external_id=external_id,
+            previous_execution_id=previous_execution_id,
+            metadata=metadata,
+            request_options=request_options,
+        )
+        return _response.data
 
     async def execute_workflow_stream(
         self,
@@ -2208,6 +1661,7 @@ class AsyncVellum:
         external_id: typing.Optional[str] = OMIT,
         event_types: typing.Optional[typing.Sequence[WorkflowExecutionEventType]] = OMIT,
         metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
+        previous_execution_id: typing.Optional[str] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[WorkflowStreamEvent]:
         """
@@ -2239,6 +1693,9 @@ class AsyncVellum:
         metadata : typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]]
             Arbitrary JSON metadata associated with this request. Can be used to capture additional monitoring data such as user id, session id, etc. for future analysis.
 
+        previous_execution_id : typing.Optional[str]
+            The ID of a previous Workflow Execution to reference for initial State loading.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -2254,6 +1711,7 @@ class AsyncVellum:
         from vellum import AsyncVellum, WorkflowRequestStringInputRequest
 
         client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
 
@@ -2262,9 +1720,13 @@ class AsyncVellum:
             response = await client.execute_workflow_stream(
                 inputs=[
                     WorkflowRequestStringInputRequest(
-                        name="name",
+                        name="x",
                         value="value",
-                    )
+                    ),
+                    WorkflowRequestStringInputRequest(
+                        name="x",
+                        value="value",
+                    ),
                 ],
             )
             async for chunk in response:
@@ -2273,81 +1735,20 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        async with self._client_wrapper.httpx_client.stream(
-            "v1/execute-workflow-stream",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "inputs": convert_and_respect_annotation_metadata(
-                    object_=inputs, annotation=typing.Sequence[WorkflowRequestInputRequest], direction="write"
-                ),
-                "expand_meta": convert_and_respect_annotation_metadata(
-                    object_=expand_meta, annotation=WorkflowExpandMetaRequest, direction="write"
-                ),
-                "workflow_deployment_id": workflow_deployment_id,
-                "workflow_deployment_name": workflow_deployment_name,
-                "release_tag": release_tag,
-                "external_id": external_id,
-                "event_types": event_types,
-                "metadata": metadata,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        async with self._raw_client.execute_workflow_stream(
+            inputs=inputs,
+            expand_meta=expand_meta,
+            workflow_deployment_id=workflow_deployment_id,
+            workflow_deployment_name=workflow_deployment_name,
+            release_tag=release_tag,
+            external_id=external_id,
+            event_types=event_types,
+            metadata=metadata,
+            previous_execution_id=previous_execution_id,
             request_options=request_options,
-            omit=OMIT,
-        ) as _response:
-            try:
-                if 200 <= _response.status_code < 300:
-                    async for _text in _response.aiter_lines():
-                        try:
-                            if len(_text) == 0:
-                                continue
-                            yield typing.cast(
-                                WorkflowStreamEvent,
-                                parse_obj_as(
-                                    type_=WorkflowStreamEvent,  # type: ignore
-                                    object_=json.loads(_text),
-                                ),
-                            )
-                        except Exception:
-                            pass
-                    return
-                await _response.aread()
-                if _response.status_code == 400:
-                    raise BadRequestError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 404:
-                    raise NotFoundError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 500:
-                    raise InternalServerError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                _response_json = _response.json()
-            except JSONDecodeError:
-                raise ApiError(status_code=_response.status_code, body=_response.text)
-            raise ApiError(status_code=_response.status_code, body=_response_json)
+        ) as r:
+            async for _chunk in r.data:
+                yield _chunk
 
     async def generate(
         self,
@@ -2393,6 +1794,7 @@ class AsyncVellum:
         from vellum import AsyncVellum, GenerateRequest
 
         client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
 
@@ -2401,87 +1803,25 @@ class AsyncVellum:
             await client.generate(
                 requests=[
                     GenerateRequest(
-                        input_values={"key": "value"},
-                    )
+                        input_values={"input_values": {"key": "value"}},
+                    ),
+                    GenerateRequest(
+                        input_values={"input_values": {"key": "value"}},
+                    ),
                 ],
             )
 
 
         asyncio.run(main())
         """
-        _response = await self._client_wrapper.httpx_client.request(
-            "v1/generate",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "deployment_id": deployment_id,
-                "deployment_name": deployment_name,
-                "requests": convert_and_respect_annotation_metadata(
-                    object_=requests, annotation=typing.Sequence[GenerateRequest], direction="write"
-                ),
-                "options": convert_and_respect_annotation_metadata(
-                    object_=options, annotation=GenerateOptionsRequest, direction="write"
-                ),
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = await self._raw_client.generate(
+            requests=requests,
+            deployment_id=deployment_id,
+            deployment_name=deployment_name,
+            options=options,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    GenerateResponse,
-                    parse_obj_as(
-                        type_=GenerateResponse,  # type: ignore
-                        object_=_response.json(),
-                    ),
-                )
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 403:
-                raise ForbiddenError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 404:
-                raise NotFoundError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 500:
-                raise InternalServerError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     async def generate_stream(
         self,
@@ -2527,6 +1867,7 @@ class AsyncVellum:
         from vellum import AsyncVellum, GenerateRequest
 
         client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
 
@@ -2535,8 +1876,11 @@ class AsyncVellum:
             response = await client.generate_stream(
                 requests=[
                     GenerateRequest(
-                        input_values={"key": "value"},
-                    )
+                        input_values={"input_values": {"key": "value"}},
+                    ),
+                    GenerateRequest(
+                        input_values={"input_values": {"key": "value"}},
+                    ),
                 ],
             )
             async for chunk in response:
@@ -2545,87 +1889,15 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        async with self._client_wrapper.httpx_client.stream(
-            "v1/generate-stream",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "deployment_id": deployment_id,
-                "deployment_name": deployment_name,
-                "requests": convert_and_respect_annotation_metadata(
-                    object_=requests, annotation=typing.Sequence[GenerateRequest], direction="write"
-                ),
-                "options": convert_and_respect_annotation_metadata(
-                    object_=options, annotation=GenerateOptionsRequest, direction="write"
-                ),
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        async with self._raw_client.generate_stream(
+            requests=requests,
+            deployment_id=deployment_id,
+            deployment_name=deployment_name,
+            options=options,
             request_options=request_options,
-            omit=OMIT,
-        ) as _response:
-            try:
-                if 200 <= _response.status_code < 300:
-                    async for _text in _response.aiter_lines():
-                        try:
-                            if len(_text) == 0:
-                                continue
-                            yield typing.cast(
-                                GenerateStreamResponse,
-                                parse_obj_as(
-                                    type_=GenerateStreamResponse,  # type: ignore
-                                    object_=json.loads(_text),
-                                ),
-                            )
-                        except Exception:
-                            pass
-                    return
-                await _response.aread()
-                if _response.status_code == 400:
-                    raise BadRequestError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 403:
-                    raise ForbiddenError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 404:
-                    raise NotFoundError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                if _response.status_code == 500:
-                    raise InternalServerError(
-                        typing.cast(
-                            typing.Optional[typing.Any],
-                            parse_obj_as(
-                                type_=typing.Optional[typing.Any],  # type: ignore
-                                object_=_response.json(),
-                            ),
-                        )
-                    )
-                _response_json = _response.json()
-            except JSONDecodeError:
-                raise ApiError(status_code=_response.status_code, body=_response.text)
-            raise ApiError(status_code=_response.status_code, body=_response_json)
+        ) as r:
+            async for _chunk in r.data:
+                yield _chunk
 
     async def search(
         self,
@@ -2672,80 +1944,28 @@ class AsyncVellum:
         from vellum import AsyncVellum
 
         client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
 
 
         async def main() -> None:
             await client.search(
-                query="query",
+                query="x",
             )
 
 
         asyncio.run(main())
         """
-        _response = await self._client_wrapper.httpx_client.request(
-            "v1/search",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "index_id": index_id,
-                "index_name": index_name,
-                "query": query,
-                "options": convert_and_respect_annotation_metadata(
-                    object_=options, annotation=SearchRequestOptionsRequest, direction="write"
-                ),
-                "document_index": document_index,
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = await self._raw_client.search(
+            query=query,
+            index_id=index_id,
+            index_name=index_name,
+            options=options,
+            document_index=document_index,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return typing.cast(
-                    SearchResponse,
-                    parse_obj_as(
-                        type_=SearchResponse,  # type: ignore
-                        object_=_response.json(),
-                    ),
-                )
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 404:
-                raise NotFoundError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 500:
-                raise InternalServerError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     async def submit_completion_actuals(
         self,
@@ -2783,72 +2003,29 @@ class AsyncVellum:
         from vellum import AsyncVellum, SubmitCompletionActualRequest
 
         client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
 
 
         async def main() -> None:
             await client.submit_completion_actuals(
-                actuals=[SubmitCompletionActualRequest()],
+                actuals=[
+                    SubmitCompletionActualRequest(),
+                    SubmitCompletionActualRequest(),
+                ],
             )
 
 
         asyncio.run(main())
         """
-        _response = await self._client_wrapper.httpx_client.request(
-            "v1/submit-completion-actuals",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "deployment_id": deployment_id,
-                "deployment_name": deployment_name,
-                "actuals": convert_and_respect_annotation_metadata(
-                    object_=actuals, annotation=typing.Sequence[SubmitCompletionActualRequest], direction="write"
-                ),
-            },
-            headers={
-                "content-type": "application/json",
-            },
+        _response = await self._raw_client.submit_completion_actuals(
+            actuals=actuals,
+            deployment_id=deployment_id,
+            deployment_name=deployment_name,
             request_options=request_options,
-            omit=OMIT,
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return
-            if _response.status_code == 400:
-                raise BadRequestError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 404:
-                raise NotFoundError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            if _response.status_code == 500:
-                raise InternalServerError(
-                    typing.cast(
-                        typing.Optional[typing.Any],
-                        parse_obj_as(
-                            type_=typing.Optional[typing.Any],  # type: ignore
-                            object_=_response.json(),
-                        ),
-                    )
-                )
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data
 
     async def submit_workflow_execution_actuals(
         self,
@@ -2888,39 +2065,23 @@ class AsyncVellum:
         from vellum import AsyncVellum, WorkflowExecutionActualStringRequest
 
         client = AsyncVellum(
+            api_version="YOUR_API_VERSION",
             api_key="YOUR_API_KEY",
         )
 
 
         async def main() -> None:
             await client.submit_workflow_execution_actuals(
-                actuals=[WorkflowExecutionActualStringRequest()],
+                actuals=[
+                    WorkflowExecutionActualStringRequest(),
+                    WorkflowExecutionActualStringRequest(),
+                ],
             )
 
 
         asyncio.run(main())
         """
-        _response = await self._client_wrapper.httpx_client.request(
-            "v1/submit-workflow-execution-actuals",
-            base_url=self._client_wrapper.get_environment().predict,
-            method="POST",
-            json={
-                "actuals": convert_and_respect_annotation_metadata(
-                    object_=actuals, annotation=typing.Sequence[SubmitWorkflowExecutionActualRequest], direction="write"
-                ),
-                "execution_id": execution_id,
-                "external_id": external_id,
-            },
-            headers={
-                "content-type": "application/json",
-            },
-            request_options=request_options,
-            omit=OMIT,
+        _response = await self._raw_client.submit_workflow_execution_actuals(
+            actuals=actuals, execution_id=execution_id, external_id=external_id, request_options=request_options
         )
-        try:
-            if 200 <= _response.status_code < 300:
-                return
-            _response_json = _response.json()
-        except JSONDecodeError:
-            raise ApiError(status_code=_response.status_code, body=_response.text)
-        raise ApiError(status_code=_response.status_code, body=_response_json)
+        return _response.data