vellum-ai 1.1.3__py3-none-any.whl → 1.1.4__py3-none-any.whl

vellum/__init__.py

@@ -3,6 +3,8 @@ from .plugins.utils import load_runtime_plugins
 
 load_runtime_plugins()
 
+# isort: skip_file
+
 from .client.types import (
     AdHocExecutePromptEvent,
     AdHocExpandMeta,

vellum/client/README.md

@@ -152,6 +152,28 @@ for chunk in response.data:
 
 ## Advanced
 
+### Access Raw Response Data
+
+The SDK provides access to raw response data, including headers, through the `.with_raw_response` property.
+The `.with_raw_response` property returns a "raw" client that can be used to access the `.headers` and `.data` attributes.
+
+```python
+from vellum import Vellum
+
+client = Vellum(
+    ...,
+)
+response = client.with_raw_response.execute_prompt(...)
+print(response.headers)  # access the response headers
+print(response.data)  # access the underlying object
+with client.ad_hoc.with_raw_response.adhoc_execute_prompt_stream(
+    ...
+) as response:
+    print(response.headers)  # access the response headers
+    for chunk in response.data:
+        print(chunk)  # access the underlying object(s)
+```
+
 ### Retries
 
 The SDK is instrumented with automatic retries with exponential backoff. A request will be retried as long

vellum/client/__init__.py

@@ -1,82 +1,62 @@
 # This file was auto-generated by Fern from our API Definition.
 
-import typing
-from .environment import VellumEnvironment
-from .types.api_version_enum import ApiVersionEnum
 import os
+import typing
+
 import httpx
 from .core.api_error import ApiError
-from .core.client_wrapper import SyncClientWrapper
-from .raw_client import RawVellum
-from .resources.ad_hoc.client import AdHocClient
-from .resources.container_images.client import ContainerImagesClient
-from .resources.deployments.client import DeploymentsClient
-from .resources.document_indexes.client import DocumentIndexesClient
-from .resources.documents.client import DocumentsClient
-from .resources.folder_entities.client import FolderEntitiesClient
-from .resources.metric_definitions.client import MetricDefinitionsClient
-from .resources.ml_models.client import MlModelsClient
-from .resources.organizations.client import OrganizationsClient
-from .resources.prompts.client import PromptsClient
-from .resources.sandboxes.client import SandboxesClient
-from .resources.test_suite_runs.client import TestSuiteRunsClient
-from .resources.test_suites.client import TestSuitesClient
-from .resources.workflow_deployments.client import WorkflowDeploymentsClient
-from .resources.workflow_executions.client import WorkflowExecutionsClient
-from .resources.workflow_sandboxes.client import WorkflowSandboxesClient
-from .resources.workflows.client import WorkflowsClient
-from .resources.workspace_secrets.client import WorkspaceSecretsClient
-from .resources.workspaces.client import WorkspacesClient
-from .types.method_enum import MethodEnum
-from .types.execute_api_request_body import ExecuteApiRequestBody
-from .types.execute_api_request_headers_value import ExecuteApiRequestHeadersValue
-from .types.execute_api_request_bearer_token import ExecuteApiRequestBearerToken
+from .core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from .core.request_options import RequestOptions
-from .types.execute_api_response import ExecuteApiResponse
+from .environment import VellumEnvironment
+from .raw_client import AsyncRawVellum, RawVellum
+from .resources.ad_hoc.client import AdHocClient, AsyncAdHocClient
+from .resources.container_images.client import AsyncContainerImagesClient, ContainerImagesClient
+from .resources.deployments.client import AsyncDeploymentsClient, DeploymentsClient
+from .resources.document_indexes.client import AsyncDocumentIndexesClient, DocumentIndexesClient
+from .resources.documents.client import AsyncDocumentsClient, DocumentsClient
+from .resources.folder_entities.client import AsyncFolderEntitiesClient, FolderEntitiesClient
+from .resources.metric_definitions.client import AsyncMetricDefinitionsClient, MetricDefinitionsClient
+from .resources.ml_models.client import AsyncMlModelsClient, MlModelsClient
+from .resources.organizations.client import AsyncOrganizationsClient, OrganizationsClient
+from .resources.prompts.client import AsyncPromptsClient, PromptsClient
+from .resources.sandboxes.client import AsyncSandboxesClient, SandboxesClient
+from .resources.test_suite_runs.client import AsyncTestSuiteRunsClient, TestSuiteRunsClient
+from .resources.test_suites.client import AsyncTestSuitesClient, TestSuitesClient
+from .resources.workflow_deployments.client import AsyncWorkflowDeploymentsClient, WorkflowDeploymentsClient
+from .resources.workflow_executions.client import AsyncWorkflowExecutionsClient, WorkflowExecutionsClient
+from .resources.workflow_sandboxes.client import AsyncWorkflowSandboxesClient, WorkflowSandboxesClient
+from .resources.workflows.client import AsyncWorkflowsClient, WorkflowsClient
+from .resources.workspace_secrets.client import AsyncWorkspaceSecretsClient, WorkspaceSecretsClient
+from .resources.workspaces.client import AsyncWorkspacesClient, WorkspacesClient
+from .types.api_version_enum import ApiVersionEnum
+from .types.code_execution_package import CodeExecutionPackage
 from .types.code_execution_runtime import CodeExecutionRuntime
 from .types.code_executor_input import CodeExecutorInput
-from .types.code_execution_package import CodeExecutionPackage
-from .types.vellum_variable_type import VellumVariableType
 from .types.code_executor_response import CodeExecutorResponse
-from .types.prompt_deployment_input_request import PromptDeploymentInputRequest
-from .types.prompt_deployment_expand_meta_request import PromptDeploymentExpandMetaRequest
-from .types.raw_prompt_execution_overrides_request import RawPromptExecutionOverridesRequest
-from .types.execute_prompt_response import ExecutePromptResponse
+from .types.execute_api_request_bearer_token import ExecuteApiRequestBearerToken
+from .types.execute_api_request_body import ExecuteApiRequestBody
+from .types.execute_api_request_headers_value import ExecuteApiRequestHeadersValue
+from .types.execute_api_response import ExecuteApiResponse
 from .types.execute_prompt_event import ExecutePromptEvent
-from .types.workflow_request_input_request import WorkflowRequestInputRequest
-from .types.workflow_expand_meta_request import WorkflowExpandMetaRequest
+from .types.execute_prompt_response import ExecutePromptResponse
 from .types.execute_workflow_response import ExecuteWorkflowResponse
-from .types.workflow_execution_event_type import WorkflowExecutionEventType
-from .types.workflow_stream_event import WorkflowStreamEvent
-from .types.generate_request import GenerateRequest
 from .types.generate_options_request import GenerateOptionsRequest
+from .types.generate_request import GenerateRequest
 from .types.generate_response import GenerateResponse
 from .types.generate_stream_response import GenerateStreamResponse
+from .types.method_enum import MethodEnum
+from .types.prompt_deployment_expand_meta_request import PromptDeploymentExpandMetaRequest
+from .types.prompt_deployment_input_request import PromptDeploymentInputRequest
+from .types.raw_prompt_execution_overrides_request import RawPromptExecutionOverridesRequest
 from .types.search_request_options_request import SearchRequestOptionsRequest
 from .types.search_response import SearchResponse
 from .types.submit_completion_actual_request import SubmitCompletionActualRequest
 from .types.submit_workflow_execution_actual_request import SubmitWorkflowExecutionActualRequest
-from .core.client_wrapper import AsyncClientWrapper
-from .raw_client import AsyncRawVellum
-from .resources.ad_hoc.client import AsyncAdHocClient
-from .resources.container_images.client import AsyncContainerImagesClient
-from .resources.deployments.client import AsyncDeploymentsClient
-from .resources.document_indexes.client import AsyncDocumentIndexesClient
-from .resources.documents.client import AsyncDocumentsClient
-from .resources.folder_entities.client import AsyncFolderEntitiesClient
-from .resources.metric_definitions.client import AsyncMetricDefinitionsClient
-from .resources.ml_models.client import AsyncMlModelsClient
-from .resources.organizations.client import AsyncOrganizationsClient
-from .resources.prompts.client import AsyncPromptsClient
-from .resources.sandboxes.client import AsyncSandboxesClient
-from .resources.test_suite_runs.client import AsyncTestSuiteRunsClient
-from .resources.test_suites.client import AsyncTestSuitesClient
-from .resources.workflow_deployments.client import AsyncWorkflowDeploymentsClient
-from .resources.workflow_executions.client import AsyncWorkflowExecutionsClient
-from .resources.workflow_sandboxes.client import AsyncWorkflowSandboxesClient
-from .resources.workflows.client import AsyncWorkflowsClient
-from .resources.workspace_secrets.client import AsyncWorkspaceSecretsClient
-from .resources.workspaces.client import AsyncWorkspacesClient
+from .types.vellum_variable_type import VellumVariableType
+from .types.workflow_execution_event_type import WorkflowExecutionEventType
+from .types.workflow_expand_meta_request import WorkflowExpandMetaRequest
+from .types.workflow_request_input_request import WorkflowRequestInputRequest
+from .types.workflow_stream_event import WorkflowStreamEvent
 
 # this is used as the default value for optional parameters
 OMIT = typing.cast(typing.Any, ...)
@@ -99,6 +79,9 @@ class Vellum:
 
     api_version : typing.Optional[ApiVersionEnum]
     api_key : str
+    headers : typing.Optional[typing.Dict[str, str]]
+        Additional headers to send with every request.
+
     timeout : typing.Optional[float]
         The timeout to be used, in seconds, for requests. By default there is no timeout set, unless a custom httpx client is used, in which case this default is not enforced.
 
@@ -124,6 +107,7 @@ class Vellum:
         environment: VellumEnvironment = VellumEnvironment.PRODUCTION,
         api_version: typing.Optional[ApiVersionEnum] = os.getenv("VELLUM_API_VERSION", "2025-07-30"),
         api_key: str,
+        headers: typing.Optional[typing.Dict[str, str]] = None,
         timeout: typing.Optional[float] = None,
         follow_redirects: typing.Optional[bool] = True,
         httpx_client: typing.Optional[httpx.Client] = None,
@@ -135,6 +119,7 @@ class Vellum:
             environment=environment,
             api_version=api_version,
             api_key=api_key,
+            headers=headers,
             httpx_client=httpx_client
             if httpx_client is not None
             else httpx.Client(timeout=_defaulted_timeout, follow_redirects=follow_redirects)
@@ -217,7 +202,7 @@ class Vellum:
             url="x",
         )
         """
-        response = self._raw_client.execute_api(
+        _response = self._raw_client.execute_api(
             url=url,
             method=method,
             body=body,
@@ -225,7 +210,7 @@ class Vellum:
             bearer_token=bearer_token,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     def execute_code(
         self,
@@ -292,7 +277,7 @@ class Vellum:
             output_type="STRING",
         )
         """
-        response = self._raw_client.execute_code(
+        _response = self._raw_client.execute_code(
             code=code,
             runtime=runtime,
             input_values=input_values,
@@ -300,7 +285,7 @@ class Vellum:
             output_type=output_type,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     def execute_prompt(
         self,
@@ -377,7 +362,7 @@ class Vellum:
             ],
         )
         """
-        response = self._raw_client.execute_prompt(
+        _response = self._raw_client.execute_prompt(
             inputs=inputs,
             prompt_deployment_id=prompt_deployment_id,
             prompt_deployment_name=prompt_deployment_name,
@@ -389,7 +374,7 @@ class Vellum:
             metadata=metadata,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     def execute_prompt_stream(
         self,
@@ -549,7 +534,7 @@ class Vellum:
             ],
         )
         """
-        response = self._raw_client.execute_workflow(
+        _response = self._raw_client.execute_workflow(
             inputs=inputs,
             expand_meta=expand_meta,
             workflow_deployment_id=workflow_deployment_id,
@@ -559,7 +544,7 @@ class Vellum:
             metadata=metadata,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     def execute_workflow_stream(
         self,
@@ -703,14 +688,14 @@ class Vellum:
             ],
         )
         """
-        response = self._raw_client.generate(
+        _response = self._raw_client.generate(
             requests=requests,
             deployment_id=deployment_id,
             deployment_name=deployment_name,
             options=options,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     def generate_stream(
         self,
@@ -829,7 +814,7 @@ class Vellum:
             query="x",
         )
         """
-        response = self._raw_client.search(
+        _response = self._raw_client.search(
             query=query,
             index_id=index_id,
             index_name=index_name,
@@ -837,7 +822,7 @@ class Vellum:
             document_index=document_index,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     def submit_completion_actuals(
         self,
@@ -880,13 +865,13 @@ class Vellum:
             actuals=[SubmitCompletionActualRequest(), SubmitCompletionActualRequest()],
         )
         """
-        response = self._raw_client.submit_completion_actuals(
+        _response = self._raw_client.submit_completion_actuals(
             actuals=actuals,
             deployment_id=deployment_id,
             deployment_name=deployment_name,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     def submit_workflow_execution_actuals(
         self,
@@ -934,10 +919,10 @@ class Vellum:
             ],
         )
         """
-        response = self._raw_client.submit_workflow_execution_actuals(
+        _response = self._raw_client.submit_workflow_execution_actuals(
             actuals=actuals, execution_id=execution_id, external_id=external_id, request_options=request_options
         )
-        return response.data
+        return _response.data
 
 
 class AsyncVellum:
@@ -957,6 +942,9 @@ class AsyncVellum:
 
     api_version : typing.Optional[ApiVersionEnum]
     api_key : str
+    headers : typing.Optional[typing.Dict[str, str]]
+        Additional headers to send with every request.
+
     timeout : typing.Optional[float]
         The timeout to be used, in seconds, for requests. By default there is no timeout set, unless a custom httpx client is used, in which case this default is not enforced.
 
@@ -982,6 +970,7 @@ class AsyncVellum:
         environment: VellumEnvironment = VellumEnvironment.PRODUCTION,
         api_version: typing.Optional[ApiVersionEnum] = os.getenv("VELLUM_API_VERSION", "2025-07-30"),
         api_key: str,
+        headers: typing.Optional[typing.Dict[str, str]] = None,
         timeout: typing.Optional[float] = None,
         follow_redirects: typing.Optional[bool] = True,
         httpx_client: typing.Optional[httpx.AsyncClient] = None,
@@ -993,6 +982,7 @@ class AsyncVellum:
             environment=environment,
             api_version=api_version,
             api_key=api_key,
+            headers=headers,
             httpx_client=httpx_client
             if httpx_client is not None
             else httpx.AsyncClient(timeout=_defaulted_timeout, follow_redirects=follow_redirects)
@@ -1083,7 +1073,7 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        response = await self._raw_client.execute_api(
+        _response = await self._raw_client.execute_api(
             url=url,
             method=method,
             body=body,
@@ -1091,7 +1081,7 @@ class AsyncVellum:
             bearer_token=bearer_token,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     async def execute_code(
         self,
@@ -1166,7 +1156,7 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        response = await self._raw_client.execute_code(
+        _response = await self._raw_client.execute_code(
             code=code,
             runtime=runtime,
             input_values=input_values,
@@ -1174,7 +1164,7 @@ class AsyncVellum:
             output_type=output_type,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     async def execute_prompt(
         self,
@@ -1259,7 +1249,7 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        response = await self._raw_client.execute_prompt(
+        _response = await self._raw_client.execute_prompt(
             inputs=inputs,
             prompt_deployment_id=prompt_deployment_id,
             prompt_deployment_name=prompt_deployment_name,
@@ -1271,7 +1261,7 @@ class AsyncVellum:
             metadata=metadata,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     async def execute_prompt_stream(
         self,
@@ -1370,8 +1360,8 @@ class AsyncVellum:
             metadata=metadata,
             request_options=request_options,
         ) as r:
-            async for data in r.data:
-                yield data
+            async for _chunk in r.data:
+                yield _chunk
 
     async def execute_workflow(
         self,
@@ -1448,7 +1438,7 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        response = await self._raw_client.execute_workflow(
+        _response = await self._raw_client.execute_workflow(
             inputs=inputs,
             expand_meta=expand_meta,
             workflow_deployment_id=workflow_deployment_id,
@@ -1458,7 +1448,7 @@ class AsyncVellum:
             metadata=metadata,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     async def execute_workflow_stream(
         self,
@@ -1552,8 +1542,8 @@ class AsyncVellum:
             metadata=metadata,
             request_options=request_options,
         ) as r:
-            async for data in r.data:
-                yield data
+            async for _chunk in r.data:
+                yield _chunk
 
     async def generate(
         self,
@@ -1619,14 +1609,14 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        response = await self._raw_client.generate(
+        _response = await self._raw_client.generate(
             requests=requests,
             deployment_id=deployment_id,
             deployment_name=deployment_name,
             options=options,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     async def generate_stream(
         self,
@@ -1701,8 +1691,8 @@ class AsyncVellum:
             options=options,
             request_options=request_options,
         ) as r:
-            async for data in r.data:
-                yield data
+            async for _chunk in r.data:
+                yield _chunk
 
     async def search(
         self,
@@ -1762,7 +1752,7 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        response = await self._raw_client.search(
+        _response = await self._raw_client.search(
             query=query,
             index_id=index_id,
             index_name=index_name,
@@ -1770,7 +1760,7 @@ class AsyncVellum:
             document_index=document_index,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     async def submit_completion_actuals(
         self,
@@ -1824,13 +1814,13 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        response = await self._raw_client.submit_completion_actuals(
+        _response = await self._raw_client.submit_completion_actuals(
             actuals=actuals,
             deployment_id=deployment_id,
             deployment_name=deployment_name,
             request_options=request_options,
         )
-        return response.data
+        return _response.data
 
     async def submit_workflow_execution_actuals(
         self,
@@ -1886,7 +1876,7 @@ class AsyncVellum:
 
         asyncio.run(main())
         """
-        response = await self._raw_client.submit_workflow_execution_actuals(
+        _response = await self._raw_client.submit_workflow_execution_actuals(
             actuals=actuals, execution_id=execution_id, external_id=external_id, request_options=request_options
         )
-        return response.data
+        return _response.data

vellum/client/core/__init__.py

@@ -1,5 +1,7 @@
 # This file was auto-generated by Fern from our API Definition.
 
+# isort: skip_file
+
 from .api_error import ApiError
 from .client_wrapper import AsyncClientWrapper, BaseClientWrapper, SyncClientWrapper
 from .datetime_utils import serialize_datetime

vellum/client/core/api_error.py

@@ -1,15 +1,23 @@
 # This file was auto-generated by Fern from our API Definition.
 
-import typing
+from typing import Any, Dict, Optional
 
 
 class ApiError(Exception):
-    status_code: typing.Optional[int]
-    body: typing.Any
+    headers: Optional[Dict[str, str]]
+    status_code: Optional[int]
+    body: Any
 
-    def __init__(self, *, status_code: typing.Optional[int] = None, body: typing.Any = None):
+    def __init__(
+        self,
+        *,
+        headers: Optional[Dict[str, str]] = None,
+        status_code: Optional[int] = None,
+        body: Any = None,
+    ) -> None:
+        self.headers = headers
         self.status_code = status_code
         self.body = body
 
     def __str__(self) -> str:
-        return f"status_code: {self.status_code}, body: {self.body}"
+        return f"headers: {self.headers}, status_code: {self.status_code}, body: {self.body}"

vellum/client/core/client_wrapper.py

@@ -15,26 +15,32 @@ class BaseClientWrapper:
         *,
         api_version: typing.Optional[ApiVersionEnum] = None,
         api_key: str,
+        headers: typing.Optional[typing.Dict[str, str]] = None,
         environment: VellumEnvironment,
         timeout: typing.Optional[float] = None,
     ):
         self._api_version = api_version
         self.api_key = api_key
+        self._headers = headers
         self._environment = environment
         self._timeout = timeout
 
     def get_headers(self) -> typing.Dict[str, str]:
         headers: typing.Dict[str, str] = {
-            "User-Agent": "vellum-ai/1.1.3",
+            "User-Agent": "vellum-ai/1.1.4",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "vellum-ai",
-            "X-Fern-SDK-Version": "1.1.3",
+            "X-Fern-SDK-Version": "1.1.4",
+            **(self.get_custom_headers() or {}),
         }
         if self._api_version is not None:
             headers["X-API-Version"] = self._api_version
         headers["X-API-KEY"] = self.api_key
         return headers
 
+    def get_custom_headers(self) -> typing.Optional[typing.Dict[str, str]]:
+        return self._headers
+
     def get_environment(self) -> VellumEnvironment:
         return self._environment
 
@@ -48,11 +54,14 @@ class SyncClientWrapper(BaseClientWrapper):
         *,
         api_version: typing.Optional[ApiVersionEnum] = None,
         api_key: str,
+        headers: typing.Optional[typing.Dict[str, str]] = None,
         environment: VellumEnvironment,
         timeout: typing.Optional[float] = None,
         httpx_client: httpx.Client,
     ):
-        super().__init__(api_version=api_version, api_key=api_key, environment=environment, timeout=timeout)
+        super().__init__(
+            api_version=api_version, api_key=api_key, headers=headers, environment=environment, timeout=timeout
+        )
         self.httpx_client = HttpClient(
             httpx_client=httpx_client, base_headers=self.get_headers, base_timeout=self.get_timeout
         )
@@ -64,11 +73,14 @@ class AsyncClientWrapper(BaseClientWrapper):
         *,
         api_version: typing.Optional[ApiVersionEnum] = None,
         api_key: str,
+        headers: typing.Optional[typing.Dict[str, str]] = None,
         environment: VellumEnvironment,
         timeout: typing.Optional[float] = None,
         httpx_client: httpx.AsyncClient,
     ):
-        super().__init__(api_version=api_version, api_key=api_key, environment=environment, timeout=timeout)
+        super().__init__(
+            api_version=api_version, api_key=api_key, headers=headers, environment=environment, timeout=timeout
+        )
         self.httpx_client = AsyncHttpClient(
             httpx_client=httpx_client, base_headers=self.get_headers, base_timeout=self.get_timeout
         )

vellum/client/core/force_multipart.py

@@ -0,0 +1,16 @@
+# This file was auto-generated by Fern from our API Definition.
+
+
+class ForceMultipartDict(dict):
+    """
+    A dictionary subclass that always evaluates to True in boolean contexts.
+
+    This is used to force multipart/form-data encoding in HTTP requests even when
+    the dictionary is empty, which would normally evaluate to False.
+    """
+
+    def __bool__(self):
+        return True
+
+
+FORCE_MULTIPART = ForceMultipartDict()