vellum-ai 0.7.11__py3-none-any.whl → 0.8.4__py3-none-any.whl

vellum/client.py

@@ -1,48 +1,57 @@
 # This file was auto-generated by Fern from our API Definition.
 
-import json
 import typing
-from json.decoder import JSONDecodeError
-
+from .environment import VellumEnvironment
 import httpx
-
-from .core.api_error import ApiError
-from .core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
-from .core.pydantic_utilities import pydantic_v1
+from .core.client_wrapper import SyncClientWrapper
+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.ml_models.client import MlModelsClient
+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.workflow_sandboxes.client import WorkflowSandboxesClient
+from .types.prompt_deployment_input_request import PromptDeploymentInputRequest
+from .types.prompt_deployment_expand_meta_request_request import PromptDeploymentExpandMetaRequestRequest
+from .types.raw_prompt_execution_overrides_request import RawPromptExecutionOverridesRequest
 from .core.request_options import RequestOptions
-from .environment import VellumEnvironment
+from .types.execute_prompt_response import ExecutePromptResponse
+from .core.pydantic_utilities import parse_obj_as
 from .errors.bad_request_error import BadRequestError
 from .errors.forbidden_error import ForbiddenError
-from .errors.internal_server_error import InternalServerError
 from .errors.not_found_error import NotFoundError
-from .resources.deployments.client import AsyncDeploymentsClient, DeploymentsClient
-from .resources.document_indexes.client import AsyncDocumentIndexesClient, DocumentIndexesClient
-from .resources.documents.client import AsyncDocumentsClient, DocumentsClient
-from .resources.folder_entities.client import AsyncFolderEntitiesClient, FolderEntitiesClient
-from .resources.ml_models.client import AsyncMlModelsClient, MlModelsClient
-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.workflow_deployments.client import AsyncWorkflowDeploymentsClient, WorkflowDeploymentsClient
-from .resources.workflow_sandboxes.client import AsyncWorkflowSandboxesClient, WorkflowSandboxesClient
+from .errors.internal_server_error import InternalServerError
+from json.decoder import JSONDecodeError
+from .core.api_error import ApiError
 from .types.execute_prompt_event import ExecutePromptEvent
-from .types.execute_prompt_response import ExecutePromptResponse
+import json
+from .types.workflow_request_input_request import WorkflowRequestInputRequest
+from .types.workflow_expand_meta_request import WorkflowExpandMetaRequest
 from .types.execute_workflow_response import ExecuteWorkflowResponse
-from .types.generate_options_request import GenerateOptionsRequest
+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_response import GenerateResponse
 from .types.generate_stream_response import GenerateStreamResponse
-from .types.prompt_deployment_expand_meta_request_request import PromptDeploymentExpandMetaRequestRequest
-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 .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
+from .core.client_wrapper import AsyncClientWrapper
+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.ml_models.client import AsyncMlModelsClient
+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.workflow_sandboxes.client import AsyncWorkflowSandboxesClient
 
 # this is used as the default value for optional parameters
 OMIT = typing.cast(typing.Any, ...)
@@ -75,7 +84,7 @@ class Vellum:
 
     Examples
     --------
-    from vellum.client import Vellum
+    from vellum import Vellum
 
     client = Vellum(
         api_key="YOUR_API_KEY",
@@ -89,7 +98,7 @@ class Vellum:
         api_key: str,
         timeout: typing.Optional[float] = None,
         follow_redirects: typing.Optional[bool] = True,
-        httpx_client: typing.Optional[httpx.Client] = None
+        httpx_client: typing.Optional[httpx.Client] = None,
     ):
         _defaulted_timeout = timeout if timeout is not None else None if httpx_client is None else None
         self._client_wrapper = SyncClientWrapper(
@@ -124,8 +133,8 @@ class Vellum:
         expand_meta: typing.Optional[PromptDeploymentExpandMetaRequestRequest] = OMIT,
         raw_overrides: typing.Optional[RawPromptExecutionOverridesRequest] = OMIT,
         expand_raw: typing.Optional[typing.Sequence[str]] = OMIT,
-        metadata: typing.Optional[typing.Dict[str, typing.Any]] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> ExecutePromptResponse:
         """
         Executes a deployed Prompt and returns the result.
@@ -156,7 +165,7 @@ class Vellum:
         expand_raw : typing.Optional[typing.Sequence[str]]
             A list of keys whose values you'd like to directly return from the JSON response of the model provider. Useful if you need lower-level info returned by model providers that Vellum would otherwise omit. Corresponding key/value pairs will be returned under the `raw` key of the API response.
 
-        metadata : typing.Optional[typing.Dict[str, typing.Any]]
+        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]
@@ -169,42 +178,18 @@ class Vellum:
 
         Examples
         --------
-        from vellum import (
-            PromptDeploymentExpandMetaRequestRequest,
-            PromptDeploymentInputRequest_String,
-            RawPromptExecutionOverridesRequest,
-        )
-        from vellum.client import Vellum
+        from vellum import StringInputRequest, Vellum
 
         client = Vellum(
             api_key="YOUR_API_KEY",
         )
         client.execute_prompt(
             inputs=[
-                PromptDeploymentInputRequest_String(
-                    name="string",
-                    value="string",
+                StringInputRequest(
+                    name="name",
+                    value="value",
                 )
             ],
-            prompt_deployment_id="string",
-            prompt_deployment_name="string",
-            release_tag="string",
-            external_id="string",
-            expand_meta=PromptDeploymentExpandMetaRequestRequest(
-                model_name=True,
-                usage=True,
-                finish_reason=True,
-                latency=True,
-                deployment_release_tag=True,
-                prompt_version_id=True,
-            ),
-            raw_overrides=RawPromptExecutionOverridesRequest(
-                body={"string": {"key": "value"}},
-                headers={"string": {"key": "value"}},
-                url="string",
-            ),
-            expand_raw=["string"],
-            metadata={"string": {"key": "value"}},
         )
         """
         _response = self._client_wrapper.httpx_client.request(
@@ -227,15 +212,53 @@ class Vellum:
         )
         try:
             if 200 <= _response.status_code < 300:
-                return pydantic_v1.parse_obj_as(ExecutePromptResponse, _response.json())  # type: ignore
+                return typing.cast(
+                    ExecutePromptResponse,
+                    parse_obj_as(
+                        type_=ExecutePromptResponse,  # type: ignore
+                        object_=_response.json(),
+                    ),
+                )
             if _response.status_code == 400:
-                raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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)
@@ -252,8 +275,8 @@ class Vellum:
         expand_meta: typing.Optional[PromptDeploymentExpandMetaRequestRequest] = OMIT,
         raw_overrides: typing.Optional[RawPromptExecutionOverridesRequest] = OMIT,
         expand_raw: typing.Optional[typing.Sequence[str]] = OMIT,
-        metadata: typing.Optional[typing.Dict[str, typing.Any]] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[ExecutePromptEvent]:
         """
         Executes a deployed Prompt and streams back the results.
@@ -284,7 +307,7 @@ class Vellum:
         expand_raw : typing.Optional[typing.Sequence[str]]
             A list of keys whose values you'd like to directly return from the JSON response of the model provider. Useful if you need lower-level info returned by model providers that Vellum would otherwise omit. Corresponding key/value pairs will be returned under the `raw` key of the API response.
 
-        metadata : typing.Optional[typing.Dict[str, typing.Any]]
+        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]
@@ -299,17 +322,17 @@ class Vellum:
         --------
         from vellum import (
             PromptDeploymentExpandMetaRequestRequest,
-            PromptDeploymentInputRequest_String,
             RawPromptExecutionOverridesRequest,
+            StringInputRequest,
+            Vellum,
         )
-        from vellum.client import Vellum
 
         client = Vellum(
             api_key="YOUR_API_KEY",
         )
         response = client.execute_prompt_stream(
             inputs=[
-                PromptDeploymentInputRequest_String(
+                StringInputRequest(
                     name="string",
                     value="string",
                 )
@@ -361,19 +384,57 @@ class Vellum:
                         try:
                             if len(_text) == 0:
                                 continue
-                            yield pydantic_v1.parse_obj_as(ExecutePromptEvent, json.loads(_text))  # type: ignore
+                            yield typing.cast(
+                                ExecutePromptEvent,
+                                parse_obj_as(
+                                    type_=ExecutePromptEvent,  # type: ignore
+                                    object_=json.loads(_text),
+                                ),
+                            )
                         except:
                             pass
                     return
                 _response.read()
                 if _response.status_code == 400:
-                    raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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)
@@ -388,7 +449,7 @@ class Vellum:
         workflow_deployment_name: typing.Optional[str] = OMIT,
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> ExecuteWorkflowResponse:
         """
         Executes a deployed Workflow and returns its outputs.
@@ -423,26 +484,18 @@ class Vellum:
 
         Examples
         --------
-        from vellum import WorkflowExpandMetaRequest, WorkflowRequestInputRequest_String
-        from vellum.client import Vellum
+        from vellum import Vellum, WorkflowRequestStringInputRequest
 
         client = Vellum(
             api_key="YOUR_API_KEY",
         )
         client.execute_workflow(
             inputs=[
-                WorkflowRequestInputRequest_String(
-                    name="string",
-                    value="string",
+                WorkflowRequestStringInputRequest(
+                    name="name",
+                    value="value",
                 )
             ],
-            expand_meta=WorkflowExpandMetaRequest(
-                usage=True,
-            ),
-            workflow_deployment_id="string",
-            workflow_deployment_name="string",
-            release_tag="string",
-            external_id="string",
         )
         """
         _response = self._client_wrapper.httpx_client.request(
@@ -462,13 +515,43 @@ class Vellum:
         )
         try:
             if 200 <= _response.status_code < 300:
-                return pydantic_v1.parse_obj_as(ExecuteWorkflowResponse, _response.json())  # type: ignore
+                return typing.cast(
+                    ExecuteWorkflowResponse,
+                    parse_obj_as(
+                        type_=ExecuteWorkflowResponse,  # type: ignore
+                        object_=_response.json(),
+                    ),
+                )
             if _response.status_code == 400:
-                raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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)
@@ -484,7 +567,7 @@ class Vellum:
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
         event_types: typing.Optional[typing.Sequence[WorkflowExecutionEventType]] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[WorkflowStreamEvent]:
         """
         Executes a deployed Workflow and streams back its results.
@@ -522,15 +605,18 @@ class Vellum:
 
         Examples
         --------
-        from vellum import WorkflowExpandMetaRequest, WorkflowRequestInputRequest_String
-        from vellum.client import Vellum
+        from vellum import (
+            Vellum,
+            WorkflowExpandMetaRequest,
+            WorkflowRequestStringInputRequest,
+        )
 
         client = Vellum(
             api_key="YOUR_API_KEY",
         )
         response = client.execute_workflow_stream(
             inputs=[
-                WorkflowRequestInputRequest_String(
+                WorkflowRequestStringInputRequest(
                     name="string",
                     value="string",
                 )
@@ -569,17 +655,47 @@ class Vellum:
                         try:
                             if len(_text) == 0:
                                 continue
-                            yield pydantic_v1.parse_obj_as(WorkflowStreamEvent, json.loads(_text))  # type: ignore
+                            yield typing.cast(
+                                WorkflowStreamEvent,
+                                parse_obj_as(
+                                    type_=WorkflowStreamEvent,  # type: ignore
+                                    object_=json.loads(_text),
+                                ),
+                            )
                         except:
                             pass
                     return
                 _response.read()
                 if _response.status_code == 400:
-                    raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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)
@@ -592,7 +708,7 @@ class Vellum:
         deployment_id: typing.Optional[str] = OMIT,
         deployment_name: typing.Optional[str] = OMIT,
         options: typing.Optional[GenerateOptionsRequest] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> GenerateResponse:
         """
         Generate a completion using a previously defined deployment.
@@ -624,8 +740,7 @@ class Vellum:
 
         Examples
         --------
-        from vellum import GenerateRequest
-        from vellum.client import Vellum
+        from vellum import GenerateRequest, Vellum
 
         client = Vellum(
             api_key="YOUR_API_KEY",
@@ -653,15 +768,53 @@ class Vellum:
         )
         try:
             if 200 <= _response.status_code < 300:
-                return pydantic_v1.parse_obj_as(GenerateResponse, _response.json())  # type: ignore
+                return typing.cast(
+                    GenerateResponse,
+                    parse_obj_as(
+                        type_=GenerateResponse,  # type: ignore
+                        object_=_response.json(),
+                    ),
+                )
             if _response.status_code == 400:
-                raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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)
@@ -674,7 +827,7 @@ class Vellum:
         deployment_id: typing.Optional[str] = OMIT,
         deployment_name: typing.Optional[str] = OMIT,
         options: typing.Optional[GenerateOptionsRequest] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[GenerateStreamResponse]:
         """
         Generate a stream of completions using a previously defined deployment.
@@ -707,12 +860,12 @@ class Vellum:
         Examples
         --------
         from vellum import (
-            ChatMessageContentRequest_String,
             ChatMessageRequest,
             GenerateOptionsRequest,
             GenerateRequest,
+            StringChatMessageContentRequest,
+            Vellum,
         )
-        from vellum.client import Vellum
 
         client = Vellum(
             api_key="YOUR_API_KEY",
@@ -727,7 +880,7 @@ class Vellum:
                         ChatMessageRequest(
                             text="string",
                             role="SYSTEM",
-                            content=ChatMessageContentRequest_String(),
+                            content=StringChatMessageContentRequest(),
                             source="string",
                         )
                     ],
@@ -760,19 +913,57 @@ class Vellum:
                         try:
                             if len(_text) == 0:
                                 continue
-                            yield pydantic_v1.parse_obj_as(GenerateStreamResponse, json.loads(_text))  # type: ignore
+                            yield typing.cast(
+                                GenerateStreamResponse,
+                                parse_obj_as(
+                                    type_=GenerateStreamResponse,  # type: ignore
+                                    object_=json.loads(_text),
+                                ),
+                            )
                         except:
                             pass
                     return
                 _response.read()
                 if _response.status_code == 400:
-                    raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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)
@@ -785,7 +976,7 @@ class Vellum:
         index_id: typing.Optional[str] = OMIT,
         index_name: typing.Optional[str] = OMIT,
         options: typing.Optional[SearchRequestOptionsRequest] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> SearchResponse:
         """
         Perform a search against a document index.
@@ -814,7 +1005,7 @@ class Vellum:
 
         Examples
         --------
-        from vellum.client import Vellum
+        from vellum import Vellum
 
         client = Vellum(
             api_key="YOUR_API_KEY",
@@ -827,19 +1018,54 @@ class Vellum:
             "v1/search",
             base_url=self._client_wrapper.get_environment().predict,
             method="POST",
-            json={"index_id": index_id, "index_name": index_name, "query": query, "options": options},
+            json={
+                "index_id": index_id,
+                "index_name": index_name,
+                "query": query,
+                "options": options,
+            },
             request_options=request_options,
             omit=OMIT,
         )
         try:
             if 200 <= _response.status_code < 300:
-                return pydantic_v1.parse_obj_as(SearchResponse, _response.json())  # type: ignore
+                return typing.cast(
+                    SearchResponse,
+                    parse_obj_as(
+                        type_=SearchResponse,  # type: ignore
+                        object_=_response.json(),
+                    ),
+                )
             if _response.status_code == 400:
-                raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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)
@@ -851,7 +1077,7 @@ class Vellum:
         actuals: typing.Sequence[SubmitCompletionActualRequest],
         deployment_id: typing.Optional[str] = OMIT,
         deployment_name: typing.Optional[str] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> None:
         """
         Used to submit feedback regarding the quality of previously generated completions.
@@ -876,8 +1102,7 @@ class Vellum:
 
         Examples
         --------
-        from vellum import SubmitCompletionActualRequest
-        from vellum.client import Vellum
+        from vellum import SubmitCompletionActualRequest, Vellum
 
         client = Vellum(
             api_key="YOUR_API_KEY",
@@ -890,7 +1115,11 @@ class Vellum:
             "v1/submit-completion-actuals",
             base_url=self._client_wrapper.get_environment().predict,
             method="POST",
-            json={"deployment_id": deployment_id, "deployment_name": deployment_name, "actuals": actuals},
+            json={
+                "deployment_id": deployment_id,
+                "deployment_name": deployment_name,
+                "actuals": actuals,
+            },
             request_options=request_options,
             omit=OMIT,
         )
@@ -898,11 +1127,35 @@ class Vellum:
             if 200 <= _response.status_code < 300:
                 return
             if _response.status_code == 400:
-                raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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)
@@ -914,7 +1167,7 @@ class Vellum:
         actuals: typing.Sequence[SubmitWorkflowExecutionActualRequest],
         execution_id: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> None:
         """
             Used to submit feedback regarding the quality of previous workflow execution and its outputs.
@@ -941,20 +1194,24 @@ class Vellum:
 
         Examples
         --------
-        from vellum.client import Vellum
+        from vellum import Vellum, WorkflowExecutionActualStringRequest
 
         client = Vellum(
             api_key="YOUR_API_KEY",
         )
         client.submit_workflow_execution_actuals(
-            actuals=[],
+            actuals=[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": actuals, "execution_id": execution_id, "external_id": external_id},
+            json={
+                "actuals": actuals,
+                "execution_id": execution_id,
+                "external_id": external_id,
+            },
             request_options=request_options,
             omit=OMIT,
         )
@@ -994,7 +1251,7 @@ class AsyncVellum:
 
     Examples
     --------
-    from vellum.client import AsyncVellum
+    from vellum import AsyncVellum
 
     client = AsyncVellum(
         api_key="YOUR_API_KEY",
@@ -1008,7 +1265,7 @@ class AsyncVellum:
         api_key: str,
         timeout: typing.Optional[float] = None,
         follow_redirects: typing.Optional[bool] = True,
-        httpx_client: typing.Optional[httpx.AsyncClient] = None
+        httpx_client: typing.Optional[httpx.AsyncClient] = None,
     ):
         _defaulted_timeout = timeout if timeout is not None else None if httpx_client is None else None
         self._client_wrapper = AsyncClientWrapper(
@@ -1043,8 +1300,8 @@ class AsyncVellum:
         expand_meta: typing.Optional[PromptDeploymentExpandMetaRequestRequest] = OMIT,
         raw_overrides: typing.Optional[RawPromptExecutionOverridesRequest] = OMIT,
         expand_raw: typing.Optional[typing.Sequence[str]] = OMIT,
-        metadata: typing.Optional[typing.Dict[str, typing.Any]] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> ExecutePromptResponse:
         """
         Executes a deployed Prompt and returns the result.
@@ -1075,7 +1332,7 @@ class AsyncVellum:
         expand_raw : typing.Optional[typing.Sequence[str]]
             A list of keys whose values you'd like to directly return from the JSON response of the model provider. Useful if you need lower-level info returned by model providers that Vellum would otherwise omit. Corresponding key/value pairs will be returned under the `raw` key of the API response.
 
-        metadata : typing.Optional[typing.Dict[str, typing.Any]]
+        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]
@@ -1090,12 +1347,7 @@ class AsyncVellum:
         --------
         import asyncio
 
-        from vellum import (
-            PromptDeploymentExpandMetaRequestRequest,
-            PromptDeploymentInputRequest_String,
-            RawPromptExecutionOverridesRequest,
-        )
-        from vellum.client import AsyncVellum
+        from vellum import AsyncVellum, StringInputRequest
 
         client = AsyncVellum(
             api_key="YOUR_API_KEY",
@@ -1105,30 +1357,11 @@ class AsyncVellum:
         async def main() -> None:
             await client.execute_prompt(
                 inputs=[
-                    PromptDeploymentInputRequest_String(
-                        name="string",
-                        value="string",
+                    StringInputRequest(
+                        name="name",
+                        value="value",
                     )
                 ],
-                prompt_deployment_id="string",
-                prompt_deployment_name="string",
-                release_tag="string",
-                external_id="string",
-                expand_meta=PromptDeploymentExpandMetaRequestRequest(
-                    model_name=True,
-                    usage=True,
-                    finish_reason=True,
-                    latency=True,
-                    deployment_release_tag=True,
-                    prompt_version_id=True,
-                ),
-                raw_overrides=RawPromptExecutionOverridesRequest(
-                    body={"string": {"key": "value"}},
-                    headers={"string": {"key": "value"}},
-                    url="string",
-                ),
-                expand_raw=["string"],
-                metadata={"string": {"key": "value"}},
             )
 
 
@@ -1154,15 +1387,53 @@ class AsyncVellum:
         )
         try:
             if 200 <= _response.status_code < 300:
-                return pydantic_v1.parse_obj_as(ExecutePromptResponse, _response.json())  # type: ignore
+                return typing.cast(
+                    ExecutePromptResponse,
+                    parse_obj_as(
+                        type_=ExecutePromptResponse,  # type: ignore
+                        object_=_response.json(),
+                    ),
+                )
             if _response.status_code == 400:
-                raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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)
@@ -1179,8 +1450,8 @@ class AsyncVellum:
         expand_meta: typing.Optional[PromptDeploymentExpandMetaRequestRequest] = OMIT,
         raw_overrides: typing.Optional[RawPromptExecutionOverridesRequest] = OMIT,
         expand_raw: typing.Optional[typing.Sequence[str]] = OMIT,
-        metadata: typing.Optional[typing.Dict[str, typing.Any]] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[ExecutePromptEvent]:
         """
         Executes a deployed Prompt and streams back the results.
@@ -1211,7 +1482,7 @@ class AsyncVellum:
         expand_raw : typing.Optional[typing.Sequence[str]]
             A list of keys whose values you'd like to directly return from the JSON response of the model provider. Useful if you need lower-level info returned by model providers that Vellum would otherwise omit. Corresponding key/value pairs will be returned under the `raw` key of the API response.
 
-        metadata : typing.Optional[typing.Dict[str, typing.Any]]
+        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]
@@ -1227,11 +1498,11 @@ class AsyncVellum:
         import asyncio
 
         from vellum import (
+            AsyncVellum,
             PromptDeploymentExpandMetaRequestRequest,
-            PromptDeploymentInputRequest_String,
             RawPromptExecutionOverridesRequest,
+            StringInputRequest,
         )
-        from vellum.client import AsyncVellum
 
         client = AsyncVellum(
             api_key="YOUR_API_KEY",
@@ -1241,7 +1512,7 @@ class AsyncVellum:
         async def main() -> None:
             response = await client.execute_prompt_stream(
                 inputs=[
-                    PromptDeploymentInputRequest_String(
+                    StringInputRequest(
                         name="string",
                         value="string",
                     )
@@ -1296,19 +1567,57 @@ class AsyncVellum:
                         try:
                             if len(_text) == 0:
                                 continue
-                            yield pydantic_v1.parse_obj_as(ExecutePromptEvent, json.loads(_text))  # type: ignore
+                            yield typing.cast(
+                                ExecutePromptEvent,
+                                parse_obj_as(
+                                    type_=ExecutePromptEvent,  # type: ignore
+                                    object_=json.loads(_text),
+                                ),
+                            )
                         except:
                             pass
                     return
                 await _response.aread()
                 if _response.status_code == 400:
-                    raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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)
@@ -1323,7 +1632,7 @@ class AsyncVellum:
         workflow_deployment_name: typing.Optional[str] = OMIT,
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> ExecuteWorkflowResponse:
         """
         Executes a deployed Workflow and returns its outputs.
@@ -1360,8 +1669,7 @@ class AsyncVellum:
         --------
         import asyncio
 
-        from vellum import WorkflowExpandMetaRequest, WorkflowRequestInputRequest_String
-        from vellum.client import AsyncVellum
+        from vellum import AsyncVellum, WorkflowRequestStringInputRequest
 
         client = AsyncVellum(
             api_key="YOUR_API_KEY",
@@ -1371,18 +1679,11 @@ class AsyncVellum:
         async def main() -> None:
             await client.execute_workflow(
                 inputs=[
-                    WorkflowRequestInputRequest_String(
-                        name="string",
-                        value="string",
+                    WorkflowRequestStringInputRequest(
+                        name="name",
+                        value="value",
                     )
                 ],
-                expand_meta=WorkflowExpandMetaRequest(
-                    usage=True,
-                ),
-                workflow_deployment_id="string",
-                workflow_deployment_name="string",
-                release_tag="string",
-                external_id="string",
             )
 
 
@@ -1405,13 +1706,43 @@ class AsyncVellum:
         )
         try:
             if 200 <= _response.status_code < 300:
-                return pydantic_v1.parse_obj_as(ExecuteWorkflowResponse, _response.json())  # type: ignore
+                return typing.cast(
+                    ExecuteWorkflowResponse,
+                    parse_obj_as(
+                        type_=ExecuteWorkflowResponse,  # type: ignore
+                        object_=_response.json(),
+                    ),
+                )
             if _response.status_code == 400:
-                raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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)
@@ -1427,7 +1758,7 @@ class AsyncVellum:
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
         event_types: typing.Optional[typing.Sequence[WorkflowExecutionEventType]] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[WorkflowStreamEvent]:
         """
         Executes a deployed Workflow and streams back its results.
@@ -1467,8 +1798,11 @@ class AsyncVellum:
         --------
         import asyncio
 
-        from vellum import WorkflowExpandMetaRequest, WorkflowRequestInputRequest_String
-        from vellum.client import AsyncVellum
+        from vellum import (
+            AsyncVellum,
+            WorkflowExpandMetaRequest,
+            WorkflowRequestStringInputRequest,
+        )
 
         client = AsyncVellum(
             api_key="YOUR_API_KEY",
@@ -1478,7 +1812,7 @@ class AsyncVellum:
         async def main() -> None:
             response = await client.execute_workflow_stream(
                 inputs=[
-                    WorkflowRequestInputRequest_String(
+                    WorkflowRequestStringInputRequest(
                         name="string",
                         value="string",
                     )
@@ -1520,17 +1854,47 @@ class AsyncVellum:
                         try:
                             if len(_text) == 0:
                                 continue
-                            yield pydantic_v1.parse_obj_as(WorkflowStreamEvent, json.loads(_text))  # type: ignore
+                            yield typing.cast(
+                                WorkflowStreamEvent,
+                                parse_obj_as(
+                                    type_=WorkflowStreamEvent,  # type: ignore
+                                    object_=json.loads(_text),
+                                ),
+                            )
                         except:
                             pass
                     return
                 await _response.aread()
                 if _response.status_code == 400:
-                    raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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)
@@ -1543,7 +1907,7 @@ class AsyncVellum:
         deployment_id: typing.Optional[str] = OMIT,
         deployment_name: typing.Optional[str] = OMIT,
         options: typing.Optional[GenerateOptionsRequest] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> GenerateResponse:
         """
         Generate a completion using a previously defined deployment.
@@ -1577,8 +1941,7 @@ class AsyncVellum:
         --------
         import asyncio
 
-        from vellum import GenerateRequest
-        from vellum.client import AsyncVellum
+        from vellum import AsyncVellum, GenerateRequest
 
         client = AsyncVellum(
             api_key="YOUR_API_KEY",
@@ -1612,15 +1975,53 @@ class AsyncVellum:
         )
         try:
             if 200 <= _response.status_code < 300:
-                return pydantic_v1.parse_obj_as(GenerateResponse, _response.json())  # type: ignore
+                return typing.cast(
+                    GenerateResponse,
+                    parse_obj_as(
+                        type_=GenerateResponse,  # type: ignore
+                        object_=_response.json(),
+                    ),
+                )
             if _response.status_code == 400:
-                raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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)
@@ -1633,7 +2034,7 @@ class AsyncVellum:
         deployment_id: typing.Optional[str] = OMIT,
         deployment_name: typing.Optional[str] = OMIT,
         options: typing.Optional[GenerateOptionsRequest] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[GenerateStreamResponse]:
         """
         Generate a stream of completions using a previously defined deployment.
@@ -1668,12 +2069,12 @@ class AsyncVellum:
         import asyncio
 
         from vellum import (
-            ChatMessageContentRequest_String,
+            AsyncVellum,
             ChatMessageRequest,
             GenerateOptionsRequest,
             GenerateRequest,
+            StringChatMessageContentRequest,
         )
-        from vellum.client import AsyncVellum
 
         client = AsyncVellum(
             api_key="YOUR_API_KEY",
@@ -1691,7 +2092,7 @@ class AsyncVellum:
                             ChatMessageRequest(
                                 text="string",
                                 role="SYSTEM",
-                                content=ChatMessageContentRequest_String(),
+                                content=StringChatMessageContentRequest(),
                                 source="string",
                             )
                         ],
@@ -1727,19 +2128,57 @@ class AsyncVellum:
                         try:
                             if len(_text) == 0:
                                 continue
-                            yield pydantic_v1.parse_obj_as(GenerateStreamResponse, json.loads(_text))  # type: ignore
+                            yield typing.cast(
+                                GenerateStreamResponse,
+                                parse_obj_as(
+                                    type_=GenerateStreamResponse,  # type: ignore
+                                    object_=json.loads(_text),
+                                ),
+                            )
                         except:
                             pass
                     return
                 await _response.aread()
                 if _response.status_code == 400:
-                    raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                    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)
@@ -1752,7 +2191,7 @@ class AsyncVellum:
         index_id: typing.Optional[str] = OMIT,
         index_name: typing.Optional[str] = OMIT,
         options: typing.Optional[SearchRequestOptionsRequest] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> SearchResponse:
         """
         Perform a search against a document index.
@@ -1783,7 +2222,7 @@ class AsyncVellum:
         --------
         import asyncio
 
-        from vellum.client import AsyncVellum
+        from vellum import AsyncVellum
 
         client = AsyncVellum(
             api_key="YOUR_API_KEY",
@@ -1802,19 +2241,54 @@ class AsyncVellum:
             "v1/search",
             base_url=self._client_wrapper.get_environment().predict,
             method="POST",
-            json={"index_id": index_id, "index_name": index_name, "query": query, "options": options},
+            json={
+                "index_id": index_id,
+                "index_name": index_name,
+                "query": query,
+                "options": options,
+            },
             request_options=request_options,
             omit=OMIT,
         )
         try:
             if 200 <= _response.status_code < 300:
-                return pydantic_v1.parse_obj_as(SearchResponse, _response.json())  # type: ignore
+                return typing.cast(
+                    SearchResponse,
+                    parse_obj_as(
+                        type_=SearchResponse,  # type: ignore
+                        object_=_response.json(),
+                    ),
+                )
             if _response.status_code == 400:
-                raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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)
@@ -1826,7 +2300,7 @@ class AsyncVellum:
         actuals: typing.Sequence[SubmitCompletionActualRequest],
         deployment_id: typing.Optional[str] = OMIT,
         deployment_name: typing.Optional[str] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> None:
         """
         Used to submit feedback regarding the quality of previously generated completions.
@@ -1853,8 +2327,7 @@ class AsyncVellum:
         --------
         import asyncio
 
-        from vellum import SubmitCompletionActualRequest
-        from vellum.client import AsyncVellum
+        from vellum import AsyncVellum, SubmitCompletionActualRequest
 
         client = AsyncVellum(
             api_key="YOUR_API_KEY",
@@ -1873,7 +2346,11 @@ class AsyncVellum:
             "v1/submit-completion-actuals",
             base_url=self._client_wrapper.get_environment().predict,
             method="POST",
-            json={"deployment_id": deployment_id, "deployment_name": deployment_name, "actuals": actuals},
+            json={
+                "deployment_id": deployment_id,
+                "deployment_name": deployment_name,
+                "actuals": actuals,
+            },
             request_options=request_options,
             omit=OMIT,
         )
@@ -1881,11 +2358,35 @@ class AsyncVellum:
             if 200 <= _response.status_code < 300:
                 return
             if _response.status_code == 400:
-                raise BadRequestError(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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(pydantic_v1.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+                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)
@@ -1897,7 +2398,7 @@ class AsyncVellum:
         actuals: typing.Sequence[SubmitWorkflowExecutionActualRequest],
         execution_id: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
-        request_options: typing.Optional[RequestOptions] = None
+        request_options: typing.Optional[RequestOptions] = None,
     ) -> None:
         """
             Used to submit feedback regarding the quality of previous workflow execution and its outputs.
@@ -1926,7 +2427,7 @@ class AsyncVellum:
         --------
         import asyncio
 
-        from vellum.client import AsyncVellum
+        from vellum import AsyncVellum, WorkflowExecutionActualStringRequest
 
         client = AsyncVellum(
             api_key="YOUR_API_KEY",
@@ -1935,7 +2436,7 @@ class AsyncVellum:
 
         async def main() -> None:
             await client.submit_workflow_execution_actuals(
-                actuals=[],
+                actuals=[WorkflowExecutionActualStringRequest()],
             )
 
 
@@ -1945,7 +2446,11 @@ class AsyncVellum:
             "v1/submit-workflow-execution-actuals",
             base_url=self._client_wrapper.get_environment().predict,
             method="POST",
-            json={"actuals": actuals, "execution_id": execution_id, "external_id": external_id},
+            json={
+                "actuals": actuals,
+                "execution_id": execution_id,
+                "external_id": external_id,
+            },
             request_options=request_options,
             omit=OMIT,
         )