vellum-ai 0.8.5__py3-none-any.whl → 0.8.7__py3-none-any.whl

vellum/client.py

@@ -4,18 +4,18 @@ import typing
 from .environment import VellumEnvironment
 import httpx
 from .core.client_wrapper import SyncClientWrapper
+from .resources.ad_hoc.client import AdHocClient
 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.prompt_deployment_expand_meta_request import PromptDeploymentExpandMetaRequest
 from .types.raw_prompt_execution_overrides_request import RawPromptExecutionOverridesRequest
 from .core.request_options import RequestOptions
 from .types.execute_prompt_response import ExecutePromptResponse
@@ -42,11 +42,11 @@ 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.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
@@ -111,11 +111,11 @@ class Vellum:
             else httpx.Client(timeout=_defaulted_timeout),
             timeout=_defaulted_timeout,
         )
+        self.ad_hoc = AdHocClient(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.folder_entities = FolderEntitiesClient(client_wrapper=self._client_wrapper)
-        self.ml_models = MlModelsClient(client_wrapper=self._client_wrapper)
         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)
@@ -130,7 +130,7 @@ class Vellum:
         prompt_deployment_name: typing.Optional[str] = OMIT,
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
-        expand_meta: typing.Optional[PromptDeploymentExpandMetaRequestRequest] = OMIT,
+        expand_meta: typing.Optional[PromptDeploymentExpandMetaRequest] = OMIT,
         raw_overrides: typing.Optional[RawPromptExecutionOverridesRequest] = OMIT,
         expand_raw: typing.Optional[typing.Sequence[str]] = OMIT,
         metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
@@ -156,7 +156,7 @@ class Vellum:
         external_id : typing.Optional[str]
             Optionally include a unique identifier for tracking purposes. Must be unique within a given Prompt Deployment.
 
-        expand_meta : typing.Optional[PromptDeploymentExpandMetaRequestRequest]
+        expand_meta : typing.Optional[PromptDeploymentExpandMetaRequest]
             An optionally specified configuration used to opt in to including additional metadata about this prompt execution in the API response. Corresponding values will be returned under the `meta` key of the API response.
 
         raw_overrides : typing.Optional[RawPromptExecutionOverridesRequest]
@@ -272,7 +272,7 @@ class Vellum:
         prompt_deployment_name: typing.Optional[str] = OMIT,
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
-        expand_meta: typing.Optional[PromptDeploymentExpandMetaRequestRequest] = OMIT,
+        expand_meta: typing.Optional[PromptDeploymentExpandMetaRequest] = OMIT,
         raw_overrides: typing.Optional[RawPromptExecutionOverridesRequest] = OMIT,
         expand_raw: typing.Optional[typing.Sequence[str]] = OMIT,
         metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
@@ -298,7 +298,7 @@ class Vellum:
         external_id : typing.Optional[str]
             Optionally include a unique identifier for tracking purposes. Must be unique within a given Prompt Deployment.
 
-        expand_meta : typing.Optional[PromptDeploymentExpandMetaRequestRequest]
+        expand_meta : typing.Optional[PromptDeploymentExpandMetaRequest]
             An optionally specified configuration used to opt in to including additional metadata about this prompt execution in the API response. Corresponding values will be returned under the `meta` key of the API response.
 
         raw_overrides : typing.Optional[RawPromptExecutionOverridesRequest]
@@ -321,7 +321,7 @@ class Vellum:
         Examples
         --------
         from vellum import (
-            PromptDeploymentExpandMetaRequestRequest,
+            PromptDeploymentExpandMetaRequest,
             RawPromptExecutionOverridesRequest,
             StringInputRequest,
             Vellum,
@@ -341,7 +341,7 @@ class Vellum:
             prompt_deployment_name="string",
             release_tag="string",
             external_id="string",
-            expand_meta=PromptDeploymentExpandMetaRequestRequest(
+            expand_meta=PromptDeploymentExpandMetaRequest(
                 model_name=True,
                 usage=True,
                 finish_reason=True,
@@ -449,6 +449,7 @@ class Vellum:
         workflow_deployment_name: typing.Optional[str] = OMIT,
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
+        metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> ExecuteWorkflowResponse:
         """
@@ -474,6 +475,9 @@ class Vellum:
         external_id : typing.Optional[str]
             Optionally include a unique identifier for tracking purposes. Must be unique for a given workflow deployment.
 
+        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.
 
@@ -509,6 +513,7 @@ class Vellum:
                 "workflow_deployment_name": workflow_deployment_name,
                 "release_tag": release_tag,
                 "external_id": external_id,
+                "metadata": metadata,
             },
             request_options=request_options,
             omit=OMIT,
@@ -567,6 +572,7 @@ class Vellum:
         release_tag: typing.Optional[str] = OMIT,
         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,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[WorkflowStreamEvent]:
         """
@@ -595,6 +601,9 @@ class Vellum:
         event_types : typing.Optional[typing.Sequence[WorkflowExecutionEventType]]
             Optionally specify which events you want to receive. Defaults to only WORKFLOW events. Note that the schema of non-WORKFLOW events is unstable and should be used with caution.
 
+        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.
 
@@ -629,6 +638,7 @@ class Vellum:
             release_tag="string",
             external_id="string",
             event_types=["NODE"],
+            metadata={"string": {"key": "value"}},
         )
         for chunk in response:
             yield chunk
@@ -645,6 +655,7 @@ class Vellum:
                 "release_tag": release_tag,
                 "external_id": external_id,
                 "event_types": event_types,
+                "metadata": metadata,
             },
             request_options=request_options,
             omit=OMIT,
@@ -1278,11 +1289,11 @@ class AsyncVellum:
             else httpx.AsyncClient(timeout=_defaulted_timeout),
             timeout=_defaulted_timeout,
         )
+        self.ad_hoc = AsyncAdHocClient(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.folder_entities = AsyncFolderEntitiesClient(client_wrapper=self._client_wrapper)
-        self.ml_models = AsyncMlModelsClient(client_wrapper=self._client_wrapper)
         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)
@@ -1297,7 +1308,7 @@ class AsyncVellum:
         prompt_deployment_name: typing.Optional[str] = OMIT,
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
-        expand_meta: typing.Optional[PromptDeploymentExpandMetaRequestRequest] = OMIT,
+        expand_meta: typing.Optional[PromptDeploymentExpandMetaRequest] = OMIT,
         raw_overrides: typing.Optional[RawPromptExecutionOverridesRequest] = OMIT,
         expand_raw: typing.Optional[typing.Sequence[str]] = OMIT,
         metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
@@ -1323,7 +1334,7 @@ class AsyncVellum:
         external_id : typing.Optional[str]
             Optionally include a unique identifier for tracking purposes. Must be unique within a given Prompt Deployment.
 
-        expand_meta : typing.Optional[PromptDeploymentExpandMetaRequestRequest]
+        expand_meta : typing.Optional[PromptDeploymentExpandMetaRequest]
             An optionally specified configuration used to opt in to including additional metadata about this prompt execution in the API response. Corresponding values will be returned under the `meta` key of the API response.
 
         raw_overrides : typing.Optional[RawPromptExecutionOverridesRequest]
@@ -1447,7 +1458,7 @@ class AsyncVellum:
         prompt_deployment_name: typing.Optional[str] = OMIT,
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
-        expand_meta: typing.Optional[PromptDeploymentExpandMetaRequestRequest] = OMIT,
+        expand_meta: typing.Optional[PromptDeploymentExpandMetaRequest] = OMIT,
         raw_overrides: typing.Optional[RawPromptExecutionOverridesRequest] = OMIT,
         expand_raw: typing.Optional[typing.Sequence[str]] = OMIT,
         metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
@@ -1473,7 +1484,7 @@ class AsyncVellum:
         external_id : typing.Optional[str]
             Optionally include a unique identifier for tracking purposes. Must be unique within a given Prompt Deployment.
 
-        expand_meta : typing.Optional[PromptDeploymentExpandMetaRequestRequest]
+        expand_meta : typing.Optional[PromptDeploymentExpandMetaRequest]
             An optionally specified configuration used to opt in to including additional metadata about this prompt execution in the API response. Corresponding values will be returned under the `meta` key of the API response.
 
         raw_overrides : typing.Optional[RawPromptExecutionOverridesRequest]
@@ -1499,7 +1510,7 @@ class AsyncVellum:
 
         from vellum import (
             AsyncVellum,
-            PromptDeploymentExpandMetaRequestRequest,
+            PromptDeploymentExpandMetaRequest,
             RawPromptExecutionOverridesRequest,
             StringInputRequest,
         )
@@ -1521,7 +1532,7 @@ class AsyncVellum:
                 prompt_deployment_name="string",
                 release_tag="string",
                 external_id="string",
-                expand_meta=PromptDeploymentExpandMetaRequestRequest(
+                expand_meta=PromptDeploymentExpandMetaRequest(
                     model_name=True,
                     usage=True,
                     finish_reason=True,
@@ -1632,6 +1643,7 @@ class AsyncVellum:
         workflow_deployment_name: typing.Optional[str] = OMIT,
         release_tag: typing.Optional[str] = OMIT,
         external_id: typing.Optional[str] = OMIT,
+        metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> ExecuteWorkflowResponse:
         """
@@ -1657,6 +1669,9 @@ class AsyncVellum:
         external_id : typing.Optional[str]
             Optionally include a unique identifier for tracking purposes. Must be unique for a given workflow deployment.
 
+        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.
 
@@ -1700,6 +1715,7 @@ class AsyncVellum:
                 "workflow_deployment_name": workflow_deployment_name,
                 "release_tag": release_tag,
                 "external_id": external_id,
+                "metadata": metadata,
             },
             request_options=request_options,
             omit=OMIT,
@@ -1758,6 +1774,7 @@ class AsyncVellum:
         release_tag: typing.Optional[str] = OMIT,
         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,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[WorkflowStreamEvent]:
         """
@@ -1786,6 +1803,9 @@ class AsyncVellum:
         event_types : typing.Optional[typing.Sequence[WorkflowExecutionEventType]]
             Optionally specify which events you want to receive. Defaults to only WORKFLOW events. Note that the schema of non-WORKFLOW events is unstable and should be used with caution.
 
+        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.
 
@@ -1825,6 +1845,7 @@ class AsyncVellum:
                 release_tag="string",
                 external_id="string",
                 event_types=["NODE"],
+                metadata={"string": {"key": "value"}},
             )
             async for chunk in response:
                 yield chunk
@@ -1844,6 +1865,7 @@ class AsyncVellum:
                 "release_tag": release_tag,
                 "external_id": external_id,
                 "event_types": event_types,
+                "metadata": metadata,
             },
             request_options=request_options,
             omit=OMIT,

vellum/core/client_wrapper.py

@@ -17,7 +17,7 @@ class BaseClientWrapper:
         headers: typing.Dict[str, str] = {
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "vellum-ai",
-            "X-Fern-SDK-Version": "0.8.5",
+            "X-Fern-SDK-Version": "0.8.7",
         }
         headers["X_API_KEY"] = self.api_key
         return headers

vellum/resources/__init__.py

@@ -1,11 +1,11 @@
 # This file was auto-generated by Fern from our API Definition.
 
 from . import (
+    ad_hoc,
     deployments,
     document_indexes,
     documents,
     folder_entities,
-    ml_models,
     sandboxes,
     test_suite_runs,
     test_suites,
@@ -20,11 +20,11 @@ __all__ = [
     "DeploymentsListRequestStatus",
     "DocumentIndexesListRequestStatus",
     "WorkflowDeploymentsListRequestStatus",
+    "ad_hoc",
     "deployments",
     "document_indexes",
     "documents",
     "folder_entities",
-    "ml_models",
     "sandboxes",
     "test_suite_runs",
     "test_suites",

vellum/resources/ad_hoc/client.py

@@ -0,0 +1,372 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+from ...core.client_wrapper import SyncClientWrapper
+from ...types.prompt_request_input_request import PromptRequestInputRequest
+from ...types.vellum_variable_request import VellumVariableRequest
+from ...types.prompt_parameters_request import PromptParametersRequest
+from ...types.prompt_block_request import PromptBlockRequest
+from ...types.ad_hoc_expand_meta_request import AdHocExpandMetaRequest
+from ...core.request_options import RequestOptions
+from ...types.ad_hoc_execute_prompt_event import AdHocExecutePromptEvent
+from ...core.pydantic_utilities import parse_obj_as
+import json
+from ...errors.bad_request_error import BadRequestError
+from ...errors.forbidden_error import ForbiddenError
+from ...errors.internal_server_error import InternalServerError
+from json.decoder import JSONDecodeError
+from ...core.api_error import ApiError
+from ...core.client_wrapper import AsyncClientWrapper
+
+# this is used as the default value for optional parameters
+OMIT = typing.cast(typing.Any, ...)
+
+
+class AdHocClient:
+    def __init__(self, *, client_wrapper: SyncClientWrapper):
+        self._client_wrapper = client_wrapper
+
+    def adhoc_execute_prompt_stream(
+        self,
+        *,
+        ml_model: str,
+        input_values: typing.Sequence[PromptRequestInputRequest],
+        input_variables: typing.Sequence[VellumVariableRequest],
+        parameters: PromptParametersRequest,
+        blocks: typing.Sequence[PromptBlockRequest],
+        expand_meta: typing.Optional[AdHocExpandMetaRequest] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> typing.Iterator[AdHocExecutePromptEvent]:
+        """
+        An internal-only endpoint that's subject to breaking changes without notice. Not intended for public use.
+
+        Parameters
+        ----------
+        ml_model : str
+
+        input_values : typing.Sequence[PromptRequestInputRequest]
+
+        input_variables : typing.Sequence[VellumVariableRequest]
+
+        parameters : PromptParametersRequest
+
+        blocks : typing.Sequence[PromptBlockRequest]
+
+        expand_meta : typing.Optional[AdHocExpandMetaRequest]
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Yields
+        ------
+        typing.Iterator[AdHocExecutePromptEvent]
+
+
+        Examples
+        --------
+        from vellum import (
+            AdHocExpandMetaRequest,
+            EphemeralPromptCacheConfigRequest,
+            JinjaPromptBlockPropertiesRequest,
+            JinjaPromptBlockRequest,
+            PromptParametersRequest,
+            PromptRequestStringInputRequest,
+            Vellum,
+            VellumVariableRequest,
+        )
+
+        client = Vellum(
+            api_key="YOUR_API_KEY",
+        )
+        response = client.ad_hoc.adhoc_execute_prompt_stream(
+            ml_model="string",
+            input_values=[
+                PromptRequestStringInputRequest(
+                    key="string",
+                    value="string",
+                )
+            ],
+            input_variables=[
+                VellumVariableRequest(
+                    id="string",
+                    key="string",
+                    type="STRING",
+                )
+            ],
+            parameters=PromptParametersRequest(
+                stop=["string"],
+                temperature=1.1,
+                max_tokens=1,
+                top_p=1.1,
+                top_k=1,
+                frequency_penalty=1.1,
+                presence_penalty=1.1,
+                logit_bias={"string": {"key": "value"}},
+                custom_parameters={"string": {"key": "value"}},
+            ),
+            blocks=[
+                JinjaPromptBlockRequest(
+                    properties=JinjaPromptBlockPropertiesRequest(
+                        template="string",
+                        template_type="STRING",
+                    ),
+                    id="string",
+                    state="ENABLED",
+                    cache_config=EphemeralPromptCacheConfigRequest(),
+                )
+            ],
+            expand_meta=AdHocExpandMetaRequest(
+                cost=True,
+                model_name=True,
+                usage=True,
+                finish_reason=True,
+            ),
+        )
+        for chunk in response:
+            yield chunk
+        """
+        with self._client_wrapper.httpx_client.stream(
+            "v1/ad-hoc/execute-prompt-stream",
+            base_url=self._client_wrapper.get_environment().default,
+            method="POST",
+            json={
+                "ml_model": ml_model,
+                "input_values": input_values,
+                "input_variables": input_variables,
+                "parameters": parameters,
+                "blocks": blocks,
+                "expand_meta": expand_meta,
+            },
+            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(
+                                AdHocExecutePromptEvent,
+                                parse_obj_as(
+                                    type_=AdHocExecutePromptEvent,  # type: ignore
+                                    object_=json.loads(_text),
+                                ),
+                            )
+                        except:
+                            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 == 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)
+
+
+class AsyncAdHocClient:
+    def __init__(self, *, client_wrapper: AsyncClientWrapper):
+        self._client_wrapper = client_wrapper
+
+    async def adhoc_execute_prompt_stream(
+        self,
+        *,
+        ml_model: str,
+        input_values: typing.Sequence[PromptRequestInputRequest],
+        input_variables: typing.Sequence[VellumVariableRequest],
+        parameters: PromptParametersRequest,
+        blocks: typing.Sequence[PromptBlockRequest],
+        expand_meta: typing.Optional[AdHocExpandMetaRequest] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> typing.AsyncIterator[AdHocExecutePromptEvent]:
+        """
+        An internal-only endpoint that's subject to breaking changes without notice. Not intended for public use.
+
+        Parameters
+        ----------
+        ml_model : str
+
+        input_values : typing.Sequence[PromptRequestInputRequest]
+
+        input_variables : typing.Sequence[VellumVariableRequest]
+
+        parameters : PromptParametersRequest
+
+        blocks : typing.Sequence[PromptBlockRequest]
+
+        expand_meta : typing.Optional[AdHocExpandMetaRequest]
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Yields
+        ------
+        typing.AsyncIterator[AdHocExecutePromptEvent]
+
+
+        Examples
+        --------
+        import asyncio
+
+        from vellum import (
+            AdHocExpandMetaRequest,
+            AsyncVellum,
+            EphemeralPromptCacheConfigRequest,
+            JinjaPromptBlockPropertiesRequest,
+            JinjaPromptBlockRequest,
+            PromptParametersRequest,
+            PromptRequestStringInputRequest,
+            VellumVariableRequest,
+        )
+
+        client = AsyncVellum(
+            api_key="YOUR_API_KEY",
+        )
+
+
+        async def main() -> None:
+            response = await client.ad_hoc.adhoc_execute_prompt_stream(
+                ml_model="string",
+                input_values=[
+                    PromptRequestStringInputRequest(
+                        key="string",
+                        value="string",
+                    )
+                ],
+                input_variables=[
+                    VellumVariableRequest(
+                        id="string",
+                        key="string",
+                        type="STRING",
+                    )
+                ],
+                parameters=PromptParametersRequest(
+                    stop=["string"],
+                    temperature=1.1,
+                    max_tokens=1,
+                    top_p=1.1,
+                    top_k=1,
+                    frequency_penalty=1.1,
+                    presence_penalty=1.1,
+                    logit_bias={"string": {"key": "value"}},
+                    custom_parameters={"string": {"key": "value"}},
+                ),
+                blocks=[
+                    JinjaPromptBlockRequest(
+                        properties=JinjaPromptBlockPropertiesRequest(
+                            template="string",
+                            template_type="STRING",
+                        ),
+                        id="string",
+                        state="ENABLED",
+                        cache_config=EphemeralPromptCacheConfigRequest(),
+                    )
+                ],
+                expand_meta=AdHocExpandMetaRequest(
+                    cost=True,
+                    model_name=True,
+                    usage=True,
+                    finish_reason=True,
+                ),
+            )
+            async for chunk in response:
+                yield chunk
+
+
+        asyncio.run(main())
+        """
+        async with self._client_wrapper.httpx_client.stream(
+            "v1/ad-hoc/execute-prompt-stream",
+            base_url=self._client_wrapper.get_environment().default,
+            method="POST",
+            json={
+                "ml_model": ml_model,
+                "input_values": input_values,
+                "input_variables": input_variables,
+                "parameters": parameters,
+                "blocks": blocks,
+                "expand_meta": expand_meta,
+            },
+            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(
+                                AdHocExecutePromptEvent,
+                                parse_obj_as(
+                                    type_=AdHocExecutePromptEvent,  # type: ignore
+                                    object_=json.loads(_text),
+                                ),
+                            )
+                        except:
+                            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 == 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)