llama-cloud 0.1.38__py3-none-any.whl → 0.1.40__py3-none-any.whl

llama_cloud/resources/retrievers/client.py

@@ -39,7 +39,6 @@ class RetrieversClient:
         name: typing.Optional[str] = None,
         project_id: typing.Optional[str] = None,
         organization_id: typing.Optional[str] = None,
-        project_id: typing.Optional[str] = None,
     ) -> typing.List[Retriever]:
         """
         List Retrievers for a project.
@@ -50,8 +49,6 @@ class RetrieversClient:
             - project_id: typing.Optional[str].
 
             - organization_id: typing.Optional[str].
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud.client import LlamaCloud
 
@@ -64,7 +61,7 @@ class RetrieversClient:
             "GET",
             urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/retrievers"),
             params=remove_none_from_dict({"name": name, "project_id": project_id, "organization_id": organization_id}),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:
@@ -83,7 +80,6 @@ class RetrieversClient:
         project_id: typing.Optional[str] = None,
         organization_id: typing.Optional[str] = None,
         request: RetrieverCreate,
-        project_id: typing.Optional[str] = None,
     ) -> Retriever:
         """
         Create a new Retriever.
@@ -94,8 +90,6 @@ class RetrieversClient:
             - organization_id: typing.Optional[str].
 
             - request: RetrieverCreate.
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud import RetrieverCreate
         from llama_cloud.client import LlamaCloud
@@ -114,7 +108,7 @@ class RetrieversClient:
             urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/retrievers"),
             params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}),
             json=jsonable_encoder(request),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:
@@ -133,7 +127,6 @@ class RetrieversClient:
         project_id: typing.Optional[str] = None,
         organization_id: typing.Optional[str] = None,
         request: RetrieverCreate,
-        project_id: typing.Optional[str] = None,
     ) -> Retriever:
         """
         Upsert a new Retriever.
@@ -144,8 +137,6 @@ class RetrieversClient:
             - organization_id: typing.Optional[str].
 
             - request: RetrieverCreate.
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud import RetrieverCreate
         from llama_cloud.client import LlamaCloud
@@ -164,7 +155,7 @@ class RetrieversClient:
             urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/retrievers"),
             params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}),
             json=jsonable_encoder(request),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:
@@ -183,7 +174,6 @@ class RetrieversClient:
         *,
         project_id: typing.Optional[str] = None,
         organization_id: typing.Optional[str] = None,
-        project_id: typing.Optional[str] = None,
     ) -> Retriever:
         """
         Get a Retriever by ID.
@@ -194,8 +184,6 @@ class RetrieversClient:
             - project_id: typing.Optional[str].
 
             - organization_id: typing.Optional[str].
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud.client import LlamaCloud
 
@@ -210,7 +198,7 @@ class RetrieversClient:
             "GET",
             urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/retrievers/{retriever_id}"),
             params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:
@@ -313,7 +301,6 @@ class RetrieversClient:
         rerank_top_n: typing.Optional[int] = OMIT,
         rerank_config: typing.Optional[ReRankConfig] = OMIT,
         query: str,
-        project_id: typing.Optional[str] = None,
     ) -> CompositeRetrievalResult:
         """
         Retrieve data using a Retriever.
@@ -332,8 +319,6 @@ class RetrieversClient:
             - rerank_config: typing.Optional[ReRankConfig]. The rerank configuration for composite retrieval.
 
             - query: str. The query to retrieve against.
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud import CompositeRetrievalMode, ReRankConfig, ReRankerType
         from llama_cloud.client import LlamaCloud
@@ -364,7 +349,7 @@ class RetrieversClient:
             ),
             params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}),
             json=jsonable_encoder(_request),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:
@@ -387,7 +372,6 @@ class RetrieversClient:
         rerank_config: typing.Optional[ReRankConfig] = OMIT,
         query: str,
         pipelines: typing.Optional[typing.List[RetrieverPipeline]] = OMIT,
-        project_id: typing.Optional[str] = None,
     ) -> CompositeRetrievalResult:
         """
         Retrieve data using specified pipelines without creating a persistent retriever.
@@ -406,8 +390,6 @@ class RetrieversClient:
             - query: str. The query to retrieve against.
 
             - pipelines: typing.Optional[typing.List[RetrieverPipeline]]. The pipelines to use for retrieval.
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud import CompositeRetrievalMode, ReRankConfig, ReRankerType
         from llama_cloud.client import LlamaCloud
@@ -437,7 +419,7 @@ class RetrieversClient:
             urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/retrievers/retrieve"),
             params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}),
             json=jsonable_encoder(_request),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:
@@ -461,7 +443,6 @@ class AsyncRetrieversClient:
         name: typing.Optional[str] = None,
         project_id: typing.Optional[str] = None,
         organization_id: typing.Optional[str] = None,
-        project_id: typing.Optional[str] = None,
     ) -> typing.List[Retriever]:
         """
         List Retrievers for a project.
@@ -472,8 +453,6 @@ class AsyncRetrieversClient:
             - project_id: typing.Optional[str].
 
             - organization_id: typing.Optional[str].
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud.client import AsyncLlamaCloud
 
@@ -486,7 +465,7 @@ class AsyncRetrieversClient:
             "GET",
             urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/retrievers"),
             params=remove_none_from_dict({"name": name, "project_id": project_id, "organization_id": organization_id}),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:
@@ -505,7 +484,6 @@ class AsyncRetrieversClient:
         project_id: typing.Optional[str] = None,
         organization_id: typing.Optional[str] = None,
         request: RetrieverCreate,
-        project_id: typing.Optional[str] = None,
     ) -> Retriever:
         """
         Create a new Retriever.
@@ -516,8 +494,6 @@ class AsyncRetrieversClient:
             - organization_id: typing.Optional[str].
 
             - request: RetrieverCreate.
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud import RetrieverCreate
         from llama_cloud.client import AsyncLlamaCloud
@@ -536,7 +512,7 @@ class AsyncRetrieversClient:
             urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/retrievers"),
             params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}),
             json=jsonable_encoder(request),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:
@@ -555,7 +531,6 @@ class AsyncRetrieversClient:
         project_id: typing.Optional[str] = None,
         organization_id: typing.Optional[str] = None,
         request: RetrieverCreate,
-        project_id: typing.Optional[str] = None,
     ) -> Retriever:
         """
         Upsert a new Retriever.
@@ -566,8 +541,6 @@ class AsyncRetrieversClient:
             - organization_id: typing.Optional[str].
 
             - request: RetrieverCreate.
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud import RetrieverCreate
         from llama_cloud.client import AsyncLlamaCloud
@@ -586,7 +559,7 @@ class AsyncRetrieversClient:
             urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/retrievers"),
             params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}),
             json=jsonable_encoder(request),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:
@@ -605,7 +578,6 @@ class AsyncRetrieversClient:
         *,
         project_id: typing.Optional[str] = None,
         organization_id: typing.Optional[str] = None,
-        project_id: typing.Optional[str] = None,
     ) -> Retriever:
         """
         Get a Retriever by ID.
@@ -616,8 +588,6 @@ class AsyncRetrieversClient:
             - project_id: typing.Optional[str].
 
             - organization_id: typing.Optional[str].
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud.client import AsyncLlamaCloud
 
@@ -632,7 +602,7 @@ class AsyncRetrieversClient:
             "GET",
             urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/retrievers/{retriever_id}"),
             params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:
@@ -735,7 +705,6 @@ class AsyncRetrieversClient:
         rerank_top_n: typing.Optional[int] = OMIT,
         rerank_config: typing.Optional[ReRankConfig] = OMIT,
         query: str,
-        project_id: typing.Optional[str] = None,
     ) -> CompositeRetrievalResult:
         """
         Retrieve data using a Retriever.
@@ -754,8 +723,6 @@ class AsyncRetrieversClient:
             - rerank_config: typing.Optional[ReRankConfig]. The rerank configuration for composite retrieval.
 
             - query: str. The query to retrieve against.
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud import CompositeRetrievalMode, ReRankConfig, ReRankerType
         from llama_cloud.client import AsyncLlamaCloud
@@ -786,7 +753,7 @@ class AsyncRetrieversClient:
             ),
             params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}),
             json=jsonable_encoder(_request),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:
@@ -809,7 +776,6 @@ class AsyncRetrieversClient:
         rerank_config: typing.Optional[ReRankConfig] = OMIT,
         query: str,
         pipelines: typing.Optional[typing.List[RetrieverPipeline]] = OMIT,
-        project_id: typing.Optional[str] = None,
     ) -> CompositeRetrievalResult:
         """
         Retrieve data using specified pipelines without creating a persistent retriever.
@@ -828,8 +794,6 @@ class AsyncRetrieversClient:
             - query: str. The query to retrieve against.
 
             - pipelines: typing.Optional[typing.List[RetrieverPipeline]]. The pipelines to use for retrieval.
-
-            - project_id: typing.Optional[str].
         ---
         from llama_cloud import CompositeRetrievalMode, ReRankConfig, ReRankerType
         from llama_cloud.client import AsyncLlamaCloud
@@ -859,7 +823,7 @@ class AsyncRetrieversClient:
             urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/retrievers/retrieve"),
             params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}),
             json=jsonable_encoder(_request),
-            headers=remove_none_from_dict({**self._client_wrapper.get_headers(), "Project-Id": project_id}),
+            headers=self._client_wrapper.get_headers(),
             timeout=60,
         )
         if 200 <= _response.status_code < 300:

llama_cloud/types/__init__.py

@@ -125,6 +125,7 @@ from .extract_job import ExtractJob
 from .extract_job_create import ExtractJobCreate
 from .extract_job_create_data_schema_override import ExtractJobCreateDataSchemaOverride
 from .extract_job_create_data_schema_override_zero_value import ExtractJobCreateDataSchemaOverrideZeroValue
+from .extract_job_create_priority import ExtractJobCreatePriority
 from .extract_mode import ExtractMode
 from .extract_models import ExtractModels
 from .extract_resultset import ExtractResultset
@@ -199,6 +200,7 @@ from .job_record_with_usage_metrics import JobRecordWithUsageMetrics
 from .l_lama_parse_transform_config import LLamaParseTransformConfig
 from .legacy_parse_job_config import LegacyParseJobConfig
 from .license_info_response import LicenseInfoResponse
+from .llama_extract_feature_availability import LlamaExtractFeatureAvailability
 from .llama_extract_mode_availability import LlamaExtractModeAvailability
 from .llama_extract_mode_availability_status import LlamaExtractModeAvailabilityStatus
 from .llama_extract_settings import LlamaExtractSettings
@@ -365,9 +367,13 @@ from .retriever import Retriever
 from .retriever_create import RetrieverCreate
 from .retriever_pipeline import RetrieverPipeline
 from .role import Role
+from .schema_generation_availability import SchemaGenerationAvailability
+from .schema_generation_availability_status import SchemaGenerationAvailabilityStatus
 from .schema_relax_mode import SchemaRelaxMode
 from .semantic_chunking_config import SemanticChunkingConfig
 from .sentence_chunking_config import SentenceChunkingConfig
+from .sparse_model_config import SparseModelConfig
+from .sparse_model_type import SparseModelType
 from .src_app_schema_chat_chat_message import SrcAppSchemaChatChatMessage
 from .status_enum import StatusEnum
 from .struct_mode import StructMode
@@ -516,6 +522,7 @@ __all__ = [
     "ExtractJobCreate",
     "ExtractJobCreateDataSchemaOverride",
     "ExtractJobCreateDataSchemaOverrideZeroValue",
+    "ExtractJobCreatePriority",
     "ExtractMode",
     "ExtractModels",
     "ExtractResultset",
@@ -588,6 +595,7 @@ __all__ = [
     "LLamaParseTransformConfig",
     "LegacyParseJobConfig",
     "LicenseInfoResponse",
+    "LlamaExtractFeatureAvailability",
     "LlamaExtractModeAvailability",
     "LlamaExtractModeAvailabilityStatus",
     "LlamaExtractSettings",
@@ -742,9 +750,13 @@ __all__ = [
     "RetrieverCreate",
     "RetrieverPipeline",
     "Role",
+    "SchemaGenerationAvailability",
+    "SchemaGenerationAvailabilityStatus",
     "SchemaRelaxMode",
     "SemanticChunkingConfig",
     "SentenceChunkingConfig",
+    "SparseModelConfig",
+    "SparseModelType",
     "SrcAppSchemaChatChatMessage",
     "StatusEnum",
     "StructMode",

llama_cloud/types/extract_job_create.py

@@ -6,6 +6,7 @@ import typing
 from ..core.datetime_utils import serialize_datetime
 from .extract_config import ExtractConfig
 from .extract_job_create_data_schema_override import ExtractJobCreateDataSchemaOverride
+from .extract_job_create_priority import ExtractJobCreatePriority
 from .webhook_configuration import WebhookConfiguration
 
 try:
@@ -22,6 +23,7 @@ class ExtractJobCreate(pydantic.BaseModel):
     Schema for creating an extraction job.
     """
 
+    priority: typing.Optional[ExtractJobCreatePriority]
     webhook_configurations: typing.Optional[typing.List[WebhookConfiguration]]
     extraction_agent_id: str = pydantic.Field(description="The id of the extraction agent")
     file_id: str = pydantic.Field(description="The id of the file")

llama_cloud/types/extract_job_create_priority.py

@@ -0,0 +1,29 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import enum
+import typing
+
+T_Result = typing.TypeVar("T_Result")
+
+
+class ExtractJobCreatePriority(str, enum.Enum):
+    LOW = "low"
+    MEDIUM = "medium"
+    HIGH = "high"
+    CRITICAL = "critical"
+
+    def visit(
+        self,
+        low: typing.Callable[[], T_Result],
+        medium: typing.Callable[[], T_Result],
+        high: typing.Callable[[], T_Result],
+        critical: typing.Callable[[], T_Result],
+    ) -> T_Result:
+        if self is ExtractJobCreatePriority.LOW:
+            return low()
+        if self is ExtractJobCreatePriority.MEDIUM:
+            return medium()
+        if self is ExtractJobCreatePriority.HIGH:
+            return high()
+        if self is ExtractJobCreatePriority.CRITICAL:
+            return critical()

llama_cloud/types/file.py

@@ -25,7 +25,7 @@ class File(pydantic.BaseModel):
     created_at: typing.Optional[dt.datetime]
     updated_at: typing.Optional[dt.datetime]
     name: str
-    external_file_id: str = pydantic.Field(description="The ID of the file in the external system")
+    external_file_id: typing.Optional[str]
     file_size: typing.Optional[int]
     file_type: typing.Optional[str]
     project_id: str = pydantic.Field(description="The ID of the project that the file belongs to")

llama_cloud/types/job_names.py

@@ -15,7 +15,6 @@ class JobNames(str, enum.Enum):
     LOAD_FILES_JOB = "load_files_job"
     PLAYGROUND_JOB = "playground_job"
     PIPELINE_MANAGED_INGESTION_JOB = "pipeline_managed_ingestion_job"
-    DATA_SOURCE_MANAGED_INGESTION_JOB = "data_source_managed_ingestion_job"
     DATA_SOURCE_UPDATE_DISPATCHER_JOB = "data_source_update_dispatcher_job"
     PIPELINE_FILE_UPDATE_DISPATCHER_JOB = "pipeline_file_update_dispatcher_job"
     PIPELINE_FILE_UPDATER_JOB = "pipeline_file_updater_job"
@@ -35,7 +34,6 @@ class JobNames(str, enum.Enum):
         load_files_job: typing.Callable[[], T_Result],
         playground_job: typing.Callable[[], T_Result],
         pipeline_managed_ingestion_job: typing.Callable[[], T_Result],
-        data_source_managed_ingestion_job: typing.Callable[[], T_Result],
         data_source_update_dispatcher_job: typing.Callable[[], T_Result],
         pipeline_file_update_dispatcher_job: typing.Callable[[], T_Result],
         pipeline_file_updater_job: typing.Callable[[], T_Result],
@@ -57,8 +55,6 @@ class JobNames(str, enum.Enum):
             return playground_job()
         if self is JobNames.PIPELINE_MANAGED_INGESTION_JOB:
             return pipeline_managed_ingestion_job()
-        if self is JobNames.DATA_SOURCE_MANAGED_INGESTION_JOB:
-            return data_source_managed_ingestion_job()
         if self is JobNames.DATA_SOURCE_UPDATE_DISPATCHER_JOB:
             return data_source_update_dispatcher_job()
         if self is JobNames.PIPELINE_FILE_UPDATE_DISPATCHER_JOB:

llama_cloud/types/llama_extract_feature_availability.py

@@ -0,0 +1,34 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import datetime as dt
+import typing
+
+from ..core.datetime_utils import serialize_datetime
+from .llama_extract_mode_availability import LlamaExtractModeAvailability
+from .schema_generation_availability import SchemaGenerationAvailability
+
+try:
+    import pydantic
+    if pydantic.__version__.startswith("1."):
+        raise ImportError
+    import pydantic.v1 as pydantic  # type: ignore
+except ImportError:
+    import pydantic  # type: ignore
+
+
+class LlamaExtractFeatureAvailability(pydantic.BaseModel):
+    schema_generation: SchemaGenerationAvailability
+    available_modes: typing.List[LlamaExtractModeAvailability]
+
+    def json(self, **kwargs: typing.Any) -> str:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().json(**kwargs_with_defaults)
+
+    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().dict(**kwargs_with_defaults)
+
+    class Config:
+        frozen = True
+        smart_union = True
+        json_encoders = {dt.datetime: serialize_datetime}

llama_cloud/types/llama_parse_parameters.py

@@ -49,6 +49,7 @@ class LlamaParseParameters(pydantic.BaseModel):
     extract_layout: typing.Optional[bool]
     high_res_ocr: typing.Optional[bool]
     html_make_all_elements_visible: typing.Optional[bool]
+    layout_aware: typing.Optional[bool]
     html_remove_navigation_elements: typing.Optional[bool]
     html_remove_fixed_elements: typing.Optional[bool]
     guess_xlsx_sheet_name: typing.Optional[bool]

llama_cloud/types/parse_job_config.py

@@ -51,6 +51,7 @@ class ParseJobConfig(pydantic.BaseModel):
     extract_layout: typing.Optional[bool]
     high_res_ocr: typing.Optional[bool]
     html_make_all_elements_visible: typing.Optional[bool]
+    layout_aware: typing.Optional[bool]
     html_remove_navigation_elements: typing.Optional[bool]
     html_remove_fixed_elements: typing.Optional[bool]
     guess_xlsx_sheet_name: typing.Optional[bool]

llama_cloud/types/pipeline.py

@@ -5,6 +5,7 @@ import typing
 
 from ..core.datetime_utils import serialize_datetime
 from .data_sink import DataSink
+from .embedding_model_config import EmbeddingModelConfig
 from .eval_execution_params import EvalExecutionParams
 from .llama_parse_parameters import LlamaParseParameters
 from .pipeline_configuration_hashes import PipelineConfigurationHashes
@@ -14,6 +15,7 @@ from .pipeline_status import PipelineStatus
 from .pipeline_transform_config import PipelineTransformConfig
 from .pipeline_type import PipelineType
 from .preset_retrieval_params import PresetRetrievalParams
+from .sparse_model_config import SparseModelConfig
 
 try:
     import pydantic
@@ -35,11 +37,13 @@ class Pipeline(pydantic.BaseModel):
     name: str
     project_id: str
     embedding_model_config_id: typing.Optional[str]
+    embedding_model_config: typing.Optional[EmbeddingModelConfig]
     pipeline_type: typing.Optional[PipelineType] = pydantic.Field(
         description="Type of pipeline. Either PLAYGROUND or MANAGED."
     )
     managed_pipeline_id: typing.Optional[str]
     embedding_config: PipelineEmbeddingConfig
+    sparse_model_config: typing.Optional[SparseModelConfig]
     config_hash: typing.Optional[PipelineConfigurationHashes]
     transform_config: typing.Optional[PipelineTransformConfig] = pydantic.Field(
         description="Configuration for the transformation."

llama_cloud/types/pipeline_create.py

@@ -12,6 +12,7 @@ from .pipeline_create_transform_config import PipelineCreateTransformConfig
 from .pipeline_metadata_config import PipelineMetadataConfig
 from .pipeline_type import PipelineType
 from .preset_retrieval_params import PresetRetrievalParams
+from .sparse_model_config import SparseModelConfig
 
 try:
     import pydantic
@@ -40,6 +41,7 @@ class PipelineCreate(pydantic.BaseModel):
     eval_parameters: typing.Optional[EvalExecutionParams] = pydantic.Field(
         description="Eval parameters for the pipeline."
     )
+    sparse_model_config: typing.Optional[SparseModelConfig]
     llama_parse_parameters: typing.Optional[LlamaParseParameters]
     status: typing.Optional[str]
     metadata_config: typing.Optional[PipelineMetadataConfig]

llama_cloud/types/pipeline_file.py

@@ -31,14 +31,14 @@ class PipelineFile(pydantic.BaseModel):
     external_file_id: typing.Optional[str]
     file_size: typing.Optional[int]
     file_type: typing.Optional[str]
-    project_id: str = pydantic.Field(description="The ID of the project that the file belongs to")
+    project_id: typing.Optional[str]
     last_modified_at: typing.Optional[dt.datetime]
-    resource_info: typing.Optional[typing.Dict[str, typing.Optional[PipelineFileResourceInfoValue]]]
-    permission_info: typing.Optional[typing.Dict[str, typing.Optional[PipelineFilePermissionInfoValue]]]
-    data_source_id: typing.Optional[str]
     file_id: typing.Optional[str]
     pipeline_id: str = pydantic.Field(description="The ID of the pipeline that the file is associated with")
+    resource_info: typing.Optional[typing.Dict[str, typing.Optional[PipelineFileResourceInfoValue]]]
+    permission_info: typing.Optional[typing.Dict[str, typing.Optional[PipelineFilePermissionInfoValue]]]
     custom_metadata: typing.Optional[typing.Dict[str, typing.Optional[PipelineFileCustomMetadataValue]]]
+    data_source_id: typing.Optional[str]
     config_hash: typing.Optional[typing.Dict[str, typing.Optional[PipelineFileConfigHashValue]]]
     indexed_page_count: typing.Optional[int]
     status: typing.Optional[PipelineFileStatus]

llama_cloud/types/schema_generation_availability.py

@@ -0,0 +1,33 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import datetime as dt
+import typing
+
+from ..core.datetime_utils import serialize_datetime
+from .schema_generation_availability_status import SchemaGenerationAvailabilityStatus
+
+try:
+    import pydantic
+    if pydantic.__version__.startswith("1."):
+        raise ImportError
+    import pydantic.v1 as pydantic  # type: ignore
+except ImportError:
+    import pydantic  # type: ignore
+
+
+class SchemaGenerationAvailability(pydantic.BaseModel):
+    model: str
+    status: SchemaGenerationAvailabilityStatus
+
+    def json(self, **kwargs: typing.Any) -> str:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().json(**kwargs_with_defaults)
+
+    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().dict(**kwargs_with_defaults)
+
+    class Config:
+        frozen = True
+        smart_union = True
+        json_encoders = {dt.datetime: serialize_datetime}

llama_cloud/types/schema_generation_availability_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 SchemaGenerationAvailabilityStatus(str, enum.Enum):
+    AVAILABLE = "available"
+    UNAVAILABLE = "unavailable"
+
+    def visit(self, available: typing.Callable[[], T_Result], unavailable: typing.Callable[[], T_Result]) -> T_Result:
+        if self is SchemaGenerationAvailabilityStatus.AVAILABLE:
+            return available()
+        if self is SchemaGenerationAvailabilityStatus.UNAVAILABLE:
+            return unavailable()

llama_cloud/types/sparse_model_config.py

@@ -0,0 +1,42 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import datetime as dt
+import typing
+
+from ..core.datetime_utils import serialize_datetime
+from .sparse_model_type import SparseModelType
+
+try:
+    import pydantic
+    if pydantic.__version__.startswith("1."):
+        raise ImportError
+    import pydantic.v1 as pydantic  # type: ignore
+except ImportError:
+    import pydantic  # type: ignore
+
+
+class SparseModelConfig(pydantic.BaseModel):
+    """
+    Configuration for sparse embedding models used in hybrid search.
+
+    This allows users to choose between Splade and BM25 models for
+    sparse retrieval in managed data sinks.
+    """
+
+    model_type: typing.Optional[SparseModelType] = pydantic.Field(
+        description="The sparse model type to use. 'auto' selects based on deployment mode (BYOC uses term frequency, Cloud uses Splade), 'splade' uses HuggingFace Splade model, 'bm25' uses Qdrant's FastEmbed BM25 model."
+    )
+    class_name: typing.Optional[str]
+
+    def json(self, **kwargs: typing.Any) -> str:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().json(**kwargs_with_defaults)
+
+    def dict(self, **kwargs: typing.Any) -> typing.Dict[str, typing.Any]:
+        kwargs_with_defaults: typing.Any = {"by_alias": True, "exclude_unset": True, **kwargs}
+        return super().dict(**kwargs_with_defaults)
+
+    class Config:
+        frozen = True
+        smart_union = True
+        json_encoders = {dt.datetime: serialize_datetime}