vellum-ai 0.1.13__py3-none-any.whl → 0.2.1__py3-none-any.whl

vellum/client.py

@@ -22,8 +22,10 @@ from .resources.model_versions.client import AsyncModelVersionsClient, ModelVers
 from .resources.registered_prompts.client import AsyncRegisteredPromptsClient, RegisteredPromptsClient
 from .resources.sandboxes.client import AsyncSandboxesClient, SandboxesClient
 from .resources.test_suites.client import AsyncTestSuitesClient, TestSuitesClient
+from .resources.workflow_deployments.client import AsyncWorkflowDeploymentsClient, WorkflowDeploymentsClient
 from .types.execute_prompt_event import ExecutePromptEvent
 from .types.execute_prompt_response import ExecutePromptResponse
+from .types.execute_workflow_response import ExecuteWorkflowResponse
 from .types.generate_options_request import GenerateOptionsRequest
 from .types.generate_request import GenerateRequest
 from .types.generate_response import GenerateResponse
@@ -69,6 +71,7 @@ class Vellum:
         self.registered_prompts = RegisteredPromptsClient(client_wrapper=self._client_wrapper)
         self.sandboxes = SandboxesClient(client_wrapper=self._client_wrapper)
         self.test_suites = TestSuitesClient(client_wrapper=self._client_wrapper)
+        self.workflow_deployments = WorkflowDeploymentsClient(client_wrapper=self._client_wrapper)
 
     def execute_prompt(
         self,
@@ -90,7 +93,7 @@ class Vellum:
         In the meantime, we recommend still using the `/generate` endpoint for prompts with function calling.
 
         Parameters:
-            - inputs: typing.List[PromptDeploymentInputRequest].
+            - inputs: typing.List[PromptDeploymentInputRequest]. The list of inputs defined in the Prompt's deployment with their corresponding values.
 
             - prompt_deployment_id: typing.Optional[str]. The ID of the Prompt Deployment. Must provide either this or prompt_deployment_name.
 
@@ -168,7 +171,7 @@ class Vellum:
         In the meantime, we recommend still using the `/generate-stream` endpoint for prompts with function calling
 
         Parameters:
-            - inputs: typing.List[PromptDeploymentInputRequest].
+            - inputs: typing.List[PromptDeploymentInputRequest]. The list of inputs defined in the Prompt's deployment with their corresponding values.
 
             - prompt_deployment_id: typing.Optional[str]. The ID of the Prompt Deployment. Must provide either this or prompt_deployment_name.
 
@@ -231,6 +234,59 @@ class Vellum:
                 raise ApiError(status_code=_response.status_code, body=_response.text)
             raise ApiError(status_code=_response.status_code, body=_response_json)
 
+    def execute_workflow(
+        self,
+        *,
+        workflow_deployment_id: typing.Optional[str] = OMIT,
+        workflow_deployment_name: typing.Optional[str] = OMIT,
+        release_tag: typing.Optional[str] = OMIT,
+        inputs: typing.List[WorkflowRequestInputRequest],
+        external_id: typing.Optional[str] = OMIT,
+    ) -> ExecuteWorkflowResponse:
+        """
+        Executes a deployed Workflow and returns its outputs.
+
+        Parameters:
+            - workflow_deployment_id: typing.Optional[str]. The ID of the Workflow Deployment. Must provide either this or workflow_deployment_name.
+
+            - workflow_deployment_name: typing.Optional[str]. The name of the Workflow Deployment. Must provide either this or workflow_deployment_id.
+
+            - release_tag: typing.Optional[str]. Optionally specify a release tag if you want to pin to a specific release of the Workflow Deployment
+
+            - inputs: typing.List[WorkflowRequestInputRequest]. The list of inputs defined in the Workflow's Deployment with their corresponding values.
+
+            - external_id: typing.Optional[str]. Optionally include a unique identifier for monitoring purposes.
+        """
+        _request: typing.Dict[str, typing.Any] = {"inputs": inputs}
+        if workflow_deployment_id is not OMIT:
+            _request["workflow_deployment_id"] = workflow_deployment_id
+        if workflow_deployment_name is not OMIT:
+            _request["workflow_deployment_name"] = workflow_deployment_name
+        if release_tag is not OMIT:
+            _request["release_tag"] = release_tag
+        if external_id is not OMIT:
+            _request["external_id"] = external_id
+        _response = self._client_wrapper.httpx_client.request(
+            "POST",
+            urllib.parse.urljoin(f"{self._client_wrapper.get_environment().default}/", "v1/execute-workflow"),
+            json=jsonable_encoder(_request),
+            headers=self._client_wrapper.get_headers(),
+            timeout=None,
+        )
+        if 200 <= _response.status_code < 300:
+            return pydantic.parse_obj_as(ExecuteWorkflowResponse, _response.json())  # type: ignore
+        if _response.status_code == 400:
+            raise BadRequestError(pydantic.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+        if _response.status_code == 404:
+            raise NotFoundError(pydantic.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+        if _response.status_code == 500:
+            raise InternalServerError(pydantic.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+        try:
+            _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)
+
     def execute_workflow_stream(
         self,
         *,
@@ -316,7 +372,13 @@ class Vellum:
 
             - options: typing.Optional[GenerateOptionsRequest]. Additional configuration that can be used to control what's included in the response.
         ---
-        from vellum import GenerateOptionsRequest, GenerateRequest, LogprobsEnum
+        from vellum import (
+            ChatMessageRequest,
+            ChatMessageRole,
+            GenerateOptionsRequest,
+            GenerateRequest,
+            LogprobsEnum,
+        )
         from vellum.client import Vellum
 
         client = Vellum(
@@ -325,7 +387,12 @@ class Vellum:
         client.generate(
             requests=[
                 GenerateRequest(
-                    input_values={},
+                    input_values={"string": {"unknown": "string", "type": "unknown"}},
+                    chat_history=[
+                        ChatMessageRequest(
+                            role=ChatMessageRole.SYSTEM,
+                        )
+                    ],
                 )
             ],
             options=GenerateOptionsRequest(
@@ -441,6 +508,41 @@ class Vellum:
             - query: str. The query to search for.
 
             - options: typing.Optional[SearchRequestOptionsRequest]. Configuration options for the search.
+        ---
+        from vellum import (
+            LogicalOperator,
+            MetadataFilterConfigRequest,
+            MetadataFilterRuleCombinator,
+            MetadataFilterRuleRequest,
+            SearchFiltersRequest,
+            SearchRequestOptionsRequest,
+            SearchResultMergingRequest,
+            SearchWeightsRequest,
+        )
+        from vellum.client import Vellum
+
+        client = Vellum(
+            api_key="YOUR_API_KEY",
+        )
+        client.search(
+            query="string",
+            options=SearchRequestOptionsRequest(
+                weights=SearchWeightsRequest(),
+                result_merging=SearchResultMergingRequest(),
+                filters=SearchFiltersRequest(
+                    metadata=MetadataFilterConfigRequest(
+                        combinator=MetadataFilterRuleCombinator.AND,
+                        rules=[
+                            MetadataFilterRuleRequest(
+                                combinator=MetadataFilterRuleCombinator.AND,
+                                operator=LogicalOperator.EQUALS,
+                            )
+                        ],
+                        operator=LogicalOperator.EQUALS,
+                    ),
+                ),
+            ),
+        )
         """
         _request: typing.Dict[str, typing.Any] = {"query": query}
         if index_id is not OMIT:
@@ -489,13 +591,14 @@ class Vellum:
 
             - actuals: typing.List[SubmitCompletionActualRequest]. Feedback regarding the quality of previously generated completions
         ---
+        from vellum import SubmitCompletionActualRequest
         from vellum.client import Vellum
 
         client = Vellum(
             api_key="YOUR_API_KEY",
         )
         client.submit_completion_actuals(
-            actuals=[],
+            actuals=[SubmitCompletionActualRequest()],
         )
         """
         _request: typing.Dict[str, typing.Any] = {"actuals": actuals}
@@ -596,6 +699,7 @@ class AsyncVellum:
         self.registered_prompts = AsyncRegisteredPromptsClient(client_wrapper=self._client_wrapper)
         self.sandboxes = AsyncSandboxesClient(client_wrapper=self._client_wrapper)
         self.test_suites = AsyncTestSuitesClient(client_wrapper=self._client_wrapper)
+        self.workflow_deployments = AsyncWorkflowDeploymentsClient(client_wrapper=self._client_wrapper)
 
     async def execute_prompt(
         self,
@@ -617,7 +721,7 @@ class AsyncVellum:
         In the meantime, we recommend still using the `/generate` endpoint for prompts with function calling.
 
         Parameters:
-            - inputs: typing.List[PromptDeploymentInputRequest].
+            - inputs: typing.List[PromptDeploymentInputRequest]. The list of inputs defined in the Prompt's deployment with their corresponding values.
 
             - prompt_deployment_id: typing.Optional[str]. The ID of the Prompt Deployment. Must provide either this or prompt_deployment_name.
 
@@ -695,7 +799,7 @@ class AsyncVellum:
         In the meantime, we recommend still using the `/generate-stream` endpoint for prompts with function calling
 
         Parameters:
-            - inputs: typing.List[PromptDeploymentInputRequest].
+            - inputs: typing.List[PromptDeploymentInputRequest]. The list of inputs defined in the Prompt's deployment with their corresponding values.
 
             - prompt_deployment_id: typing.Optional[str]. The ID of the Prompt Deployment. Must provide either this or prompt_deployment_name.
 
@@ -758,6 +862,59 @@ class AsyncVellum:
                 raise ApiError(status_code=_response.status_code, body=_response.text)
             raise ApiError(status_code=_response.status_code, body=_response_json)
 
+    async def execute_workflow(
+        self,
+        *,
+        workflow_deployment_id: typing.Optional[str] = OMIT,
+        workflow_deployment_name: typing.Optional[str] = OMIT,
+        release_tag: typing.Optional[str] = OMIT,
+        inputs: typing.List[WorkflowRequestInputRequest],
+        external_id: typing.Optional[str] = OMIT,
+    ) -> ExecuteWorkflowResponse:
+        """
+        Executes a deployed Workflow and returns its outputs.
+
+        Parameters:
+            - workflow_deployment_id: typing.Optional[str]. The ID of the Workflow Deployment. Must provide either this or workflow_deployment_name.
+
+            - workflow_deployment_name: typing.Optional[str]. The name of the Workflow Deployment. Must provide either this or workflow_deployment_id.
+
+            - release_tag: typing.Optional[str]. Optionally specify a release tag if you want to pin to a specific release of the Workflow Deployment
+
+            - inputs: typing.List[WorkflowRequestInputRequest]. The list of inputs defined in the Workflow's Deployment with their corresponding values.
+
+            - external_id: typing.Optional[str]. Optionally include a unique identifier for monitoring purposes.
+        """
+        _request: typing.Dict[str, typing.Any] = {"inputs": inputs}
+        if workflow_deployment_id is not OMIT:
+            _request["workflow_deployment_id"] = workflow_deployment_id
+        if workflow_deployment_name is not OMIT:
+            _request["workflow_deployment_name"] = workflow_deployment_name
+        if release_tag is not OMIT:
+            _request["release_tag"] = release_tag
+        if external_id is not OMIT:
+            _request["external_id"] = external_id
+        _response = await self._client_wrapper.httpx_client.request(
+            "POST",
+            urllib.parse.urljoin(f"{self._client_wrapper.get_environment().default}/", "v1/execute-workflow"),
+            json=jsonable_encoder(_request),
+            headers=self._client_wrapper.get_headers(),
+            timeout=None,
+        )
+        if 200 <= _response.status_code < 300:
+            return pydantic.parse_obj_as(ExecuteWorkflowResponse, _response.json())  # type: ignore
+        if _response.status_code == 400:
+            raise BadRequestError(pydantic.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+        if _response.status_code == 404:
+            raise NotFoundError(pydantic.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+        if _response.status_code == 500:
+            raise InternalServerError(pydantic.parse_obj_as(typing.Any, _response.json()))  # type: ignore
+        try:
+            _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)
+
     async def execute_workflow_stream(
         self,
         *,
@@ -843,7 +1000,13 @@ class AsyncVellum:
 
             - options: typing.Optional[GenerateOptionsRequest]. Additional configuration that can be used to control what's included in the response.
         ---
-        from vellum import GenerateOptionsRequest, GenerateRequest, LogprobsEnum
+        from vellum import (
+            ChatMessageRequest,
+            ChatMessageRole,
+            GenerateOptionsRequest,
+            GenerateRequest,
+            LogprobsEnum,
+        )
         from vellum.client import AsyncVellum
 
         client = AsyncVellum(
@@ -852,7 +1015,12 @@ class AsyncVellum:
         await client.generate(
             requests=[
                 GenerateRequest(
-                    input_values={},
+                    input_values={"string": {"unknown": "string", "type": "unknown"}},
+                    chat_history=[
+                        ChatMessageRequest(
+                            role=ChatMessageRole.SYSTEM,
+                        )
+                    ],
                 )
             ],
             options=GenerateOptionsRequest(
@@ -968,6 +1136,41 @@ class AsyncVellum:
             - query: str. The query to search for.
 
             - options: typing.Optional[SearchRequestOptionsRequest]. Configuration options for the search.
+        ---
+        from vellum import (
+            LogicalOperator,
+            MetadataFilterConfigRequest,
+            MetadataFilterRuleCombinator,
+            MetadataFilterRuleRequest,
+            SearchFiltersRequest,
+            SearchRequestOptionsRequest,
+            SearchResultMergingRequest,
+            SearchWeightsRequest,
+        )
+        from vellum.client import AsyncVellum
+
+        client = AsyncVellum(
+            api_key="YOUR_API_KEY",
+        )
+        await client.search(
+            query="string",
+            options=SearchRequestOptionsRequest(
+                weights=SearchWeightsRequest(),
+                result_merging=SearchResultMergingRequest(),
+                filters=SearchFiltersRequest(
+                    metadata=MetadataFilterConfigRequest(
+                        combinator=MetadataFilterRuleCombinator.AND,
+                        rules=[
+                            MetadataFilterRuleRequest(
+                                combinator=MetadataFilterRuleCombinator.AND,
+                                operator=LogicalOperator.EQUALS,
+                            )
+                        ],
+                        operator=LogicalOperator.EQUALS,
+                    ),
+                ),
+            ),
+        )
         """
         _request: typing.Dict[str, typing.Any] = {"query": query}
         if index_id is not OMIT:
@@ -1016,13 +1219,14 @@ class AsyncVellum:
 
             - actuals: typing.List[SubmitCompletionActualRequest]. Feedback regarding the quality of previously generated completions
         ---
+        from vellum import SubmitCompletionActualRequest
         from vellum.client import AsyncVellum
 
         client = AsyncVellum(
             api_key="YOUR_API_KEY",
         )
         await client.submit_completion_actuals(
-            actuals=[],
+            actuals=[SubmitCompletionActualRequest()],
         )
         """
         _request: typing.Dict[str, typing.Any] = {"actuals": actuals}

vellum/core/client_wrapper.py

@@ -16,7 +16,7 @@ class BaseClientWrapper:
         headers: typing.Dict[str, str] = {
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "vellum-ai",
-            "X-Fern-SDK-Version": "v0.1.13",
+            "X-Fern-SDK-Version": "v0.2.1",
         }
         headers["X_API_KEY"] = self.api_key
         return headers

vellum/resources/__init__.py

@@ -1,8 +1,21 @@
 # This file was auto-generated by Fern from our API Definition.
 
-from . import deployments, document_indexes, documents, model_versions, registered_prompts, sandboxes, test_suites
+from . import (
+    deployments,
+    document_indexes,
+    documents,
+    model_versions,
+    registered_prompts,
+    sandboxes,
+    test_suites,
+    workflow_deployments,
+)
+from .deployments import DeploymentsListRequestStatus
+from .workflow_deployments import WorkflowDeploymentsListRequestStatus
 
 __all__ = [
+    "DeploymentsListRequestStatus",
+    "WorkflowDeploymentsListRequestStatus",
     "deployments",
     "document_indexes",
     "documents",
@@ -10,4 +23,5 @@ __all__ = [
     "registered_prompts",
     "sandboxes",
     "test_suites",
+    "workflow_deployments",
 ]

vellum/resources/deployments/__init__.py

@@ -1,2 +1,5 @@
 # This file was auto-generated by Fern from our API Definition.
 
+from .types import DeploymentsListRequestStatus
+
+__all__ = ["DeploymentsListRequestStatus"]

vellum/resources/deployments/client.py

@@ -7,9 +7,12 @@ from json.decoder import JSONDecodeError
 from ...core.api_error import ApiError
 from ...core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ...core.jsonable_encoder import jsonable_encoder
+from ...core.remove_none_from_dict import remove_none_from_dict
 from ...types.deployment_provider_payload_response import DeploymentProviderPayloadResponse
 from ...types.deployment_read import DeploymentRead
+from ...types.paginated_slim_deployment_read_list import PaginatedSlimDeploymentReadList
 from ...types.prompt_deployment_input_request import PromptDeploymentInputRequest
+from .types.deployments_list_request_status import DeploymentsListRequestStatus
 
 try:
     import pydantic.v1 as pydantic  # type: ignore
@@ -24,12 +27,66 @@ class DeploymentsClient:
     def __init__(self, *, client_wrapper: SyncClientWrapper):
         self._client_wrapper = client_wrapper
 
+    def list(
+        self,
+        *,
+        limit: typing.Optional[int] = None,
+        offset: typing.Optional[int] = None,
+        ordering: typing.Optional[str] = None,
+        status: typing.Optional[DeploymentsListRequestStatus] = None,
+    ) -> PaginatedSlimDeploymentReadList:
+        """
+        Parameters:
+            - limit: typing.Optional[int]. Number of results to return per page.
+
+            - offset: typing.Optional[int]. The initial index from which to return the results.
+
+            - ordering: typing.Optional[str]. Which field to use when ordering the results.
+
+            - status: typing.Optional[DeploymentsListRequestStatus]. The current status of the deployment
+
+                                                                     - `ACTIVE` - Active
+                                                                     - `ARCHIVED` - Archived---
+        from vellum import DeploymentsListRequestStatus
+        from vellum.client import Vellum
+
+        client = Vellum(
+            api_key="YOUR_API_KEY",
+        )
+        client.deployments.list(
+            status=DeploymentsListRequestStatus.ACTIVE,
+        )
+        """
+        _response = self._client_wrapper.httpx_client.request(
+            "GET",
+            urllib.parse.urljoin(f"{self._client_wrapper.get_environment().default}/", "v1/deployments"),
+            params=remove_none_from_dict({"limit": limit, "offset": offset, "ordering": ordering, "status": status}),
+            headers=self._client_wrapper.get_headers(),
+            timeout=None,
+        )
+        if 200 <= _response.status_code < 300:
+            return pydantic.parse_obj_as(PaginatedSlimDeploymentReadList, _response.json())  # type: ignore
+        try:
+            _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)
+
     def retrieve(self, id: str) -> DeploymentRead:
         """
         Used to retrieve a deployment given its ID or name.
 
         Parameters:
             - id: str. Either the Deployment's ID or its unique name
+        ---
+        from vellum.client import Vellum
+
+        client = Vellum(
+            api_key="YOUR_API_KEY",
+        )
+        client.deployments.retrieve(
+            id="string",
+        )
         """
         _response = self._client_wrapper.httpx_client.request(
             "GET",
@@ -58,7 +115,7 @@ class DeploymentsClient:
 
             - deployment_name: typing.Optional[str]. The name of the deployment. Must provide either this or deployment_id.
 
-            - inputs: typing.List[PromptDeploymentInputRequest].
+            - inputs: typing.List[PromptDeploymentInputRequest]. The list of inputs defined in the Prompt's deployment with their corresponding values.
         ---
         from vellum.client import Vellum
 
@@ -96,12 +153,66 @@ class AsyncDeploymentsClient:
     def __init__(self, *, client_wrapper: AsyncClientWrapper):
         self._client_wrapper = client_wrapper
 
+    async def list(
+        self,
+        *,
+        limit: typing.Optional[int] = None,
+        offset: typing.Optional[int] = None,
+        ordering: typing.Optional[str] = None,
+        status: typing.Optional[DeploymentsListRequestStatus] = None,
+    ) -> PaginatedSlimDeploymentReadList:
+        """
+        Parameters:
+            - limit: typing.Optional[int]. Number of results to return per page.
+
+            - offset: typing.Optional[int]. The initial index from which to return the results.
+
+            - ordering: typing.Optional[str]. Which field to use when ordering the results.
+
+            - status: typing.Optional[DeploymentsListRequestStatus]. The current status of the deployment
+
+                                                                     - `ACTIVE` - Active
+                                                                     - `ARCHIVED` - Archived---
+        from vellum import DeploymentsListRequestStatus
+        from vellum.client import AsyncVellum
+
+        client = AsyncVellum(
+            api_key="YOUR_API_KEY",
+        )
+        await client.deployments.list(
+            status=DeploymentsListRequestStatus.ACTIVE,
+        )
+        """
+        _response = await self._client_wrapper.httpx_client.request(
+            "GET",
+            urllib.parse.urljoin(f"{self._client_wrapper.get_environment().default}/", "v1/deployments"),
+            params=remove_none_from_dict({"limit": limit, "offset": offset, "ordering": ordering, "status": status}),
+            headers=self._client_wrapper.get_headers(),
+            timeout=None,
+        )
+        if 200 <= _response.status_code < 300:
+            return pydantic.parse_obj_as(PaginatedSlimDeploymentReadList, _response.json())  # type: ignore
+        try:
+            _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)
+
     async def retrieve(self, id: str) -> DeploymentRead:
         """
         Used to retrieve a deployment given its ID or name.
 
         Parameters:
             - id: str. Either the Deployment's ID or its unique name
+        ---
+        from vellum.client import AsyncVellum
+
+        client = AsyncVellum(
+            api_key="YOUR_API_KEY",
+        )
+        await client.deployments.retrieve(
+            id="string",
+        )
         """
         _response = await self._client_wrapper.httpx_client.request(
             "GET",
@@ -130,7 +241,7 @@ class AsyncDeploymentsClient:
 
             - deployment_name: typing.Optional[str]. The name of the deployment. Must provide either this or deployment_id.
 
-            - inputs: typing.List[PromptDeploymentInputRequest].
+            - inputs: typing.List[PromptDeploymentInputRequest]. The list of inputs defined in the Prompt's deployment with their corresponding values.
         ---
         from vellum.client import AsyncVellum
 

vellum/resources/deployments/types/__init__.py

@@ -0,0 +1,5 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from .deployments_list_request_status import DeploymentsListRequestStatus
+
+__all__ = ["DeploymentsListRequestStatus"]

vellum/resources/deployments/types/deployments_list_request_status.py

@@ -0,0 +1,17 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import enum
+import typing
+
+T_Result = typing.TypeVar("T_Result")
+
+
+class DeploymentsListRequestStatus(str, enum.Enum):
+    ACTIVE = "ACTIVE"
+    ARCHIVED = "ARCHIVED"
+
+    def visit(self, active: typing.Callable[[], T_Result], archived: typing.Callable[[], T_Result]) -> T_Result:
+        if self is DeploymentsListRequestStatus.ACTIVE:
+            return active()
+        if self is DeploymentsListRequestStatus.ARCHIVED:
+            return archived()