Skip to content
LlamaIndex OSS Documentation
API Reference

Pipelines

Search Pipelines
pipelines.list(PipelineListParams**kwargs) -> PipelineListResponse
GET/api/v1/pipelines
Create Pipeline
pipelines.create(PipelineCreateParams**kwargs) -> Pipeline
POST/api/v1/pipelines
Get Pipeline
pipelines.get(strpipeline_id) -> Pipeline
GET/api/v1/pipelines/{pipeline_id}
Update Existing Pipeline
pipelines.update(strpipeline_id, PipelineUpdateParams**kwargs) -> Pipeline
PUT/api/v1/pipelines/{pipeline_id}
Delete Pipeline
pipelines.delete(strpipeline_id)
DELETE/api/v1/pipelines/{pipeline_id}
Get Pipeline Status
pipelines.get_status(strpipeline_id, PipelineGetStatusParams**kwargs) -> ManagedIngestionStatusResponse
GET/api/v1/pipelines/{pipeline_id}/status
Upsert Pipeline
pipelines.upsert(PipelineUpsertParams**kwargs) -> Pipeline
PUT/api/v1/pipelines
Run Search
pipelines.retrieve(strpipeline_id, PipelineRetrieveParams**kwargs) -> PipelineRetrieveResponse
POST/api/v1/pipelines/{pipeline_id}/retrieve
ModelsExpand Collapse
class AdvancedModeTransformConfig: …
chunking_config: Optional[ChunkingConfig]

Configuration for the chunking.

One of the following:
class ChunkingConfigNoneChunkingConfig: …
mode: Optional[Literal["none"]]
class ChunkingConfigCharacterChunkingConfig: …
chunk_overlap: Optional[int]
chunk_size: Optional[int]
mode: Optional[Literal["character"]]
class ChunkingConfigTokenChunkingConfig: …
chunk_overlap: Optional[int]
chunk_size: Optional[int]
mode: Optional[Literal["token"]]
separator: Optional[str]
class ChunkingConfigSentenceChunkingConfig: …
chunk_overlap: Optional[int]
chunk_size: Optional[int]
mode: Optional[Literal["sentence"]]
paragraph_separator: Optional[str]
separator: Optional[str]
class ChunkingConfigSemanticChunkingConfig: …
breakpoint_percentile_threshold: Optional[int]
buffer_size: Optional[int]
mode: Optional[Literal["semantic"]]
mode: Optional[Literal["advanced"]]
segmentation_config: Optional[SegmentationConfig]

Configuration for the segmentation.

One of the following:
class SegmentationConfigNoneSegmentationConfig: …
mode: Optional[Literal["none"]]
class SegmentationConfigPageSegmentationConfig: …
mode: Optional[Literal["page"]]
page_separator: Optional[str]
class SegmentationConfigElementSegmentationConfig: …
mode: Optional[Literal["element"]]
class AutoTransformConfig: …
chunk_overlap: Optional[int]

Chunk overlap for the transformation.

chunk_size: Optional[int]

Chunk size for the transformation.

exclusiveMinimum0
mode: Optional[Literal["auto"]]
class AzureOpenAIEmbedding: …
additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for Azure deployment.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for Azure OpenAI API.

azure_deployment: Optional[str]

The Azure deployment to use.

azure_endpoint: Optional[str]

The Azure endpoint to use.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
class AzureOpenAIEmbeddingConfig: …
component: Optional[AzureOpenAIEmbedding]

Configuration for the Azure OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for Azure deployment.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for Azure OpenAI API.

azure_deployment: Optional[str]

The Azure deployment to use.

azure_endpoint: Optional[str]

The Azure endpoint to use.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["AZURE_EMBEDDING"]]

Type of the embedding model.

class BedrockEmbedding: …
additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the bedrock client.

aws_access_key_id: Optional[str]

AWS Access Key ID to use

aws_secret_access_key: Optional[str]

AWS Secret Access Key to use

aws_session_token: Optional[str]

AWS Session Token to use

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

The maximum number of API retries.

exclusiveMinimum0
model_name: Optional[str]

The modelId of the Bedrock model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

profile_name: Optional[str]

The name of aws profile to use. If not given, then the default profile is used.

region_name: Optional[str]

AWS region name to use. Uses region configured in AWS CLI if not passed

timeout: Optional[float]

The timeout for the Bedrock API request in seconds. It will be used for both connect and read timeouts.

class BedrockEmbeddingConfig: …
component: Optional[BedrockEmbedding]

Configuration for the Bedrock embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the bedrock client.

aws_access_key_id: Optional[str]

AWS Access Key ID to use

aws_secret_access_key: Optional[str]

AWS Secret Access Key to use

aws_session_token: Optional[str]

AWS Session Token to use

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

The maximum number of API retries.

exclusiveMinimum0
model_name: Optional[str]

The modelId of the Bedrock model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

profile_name: Optional[str]

The name of aws profile to use. If not given, then the default profile is used.

region_name: Optional[str]

AWS region name to use. Uses region configured in AWS CLI if not passed

timeout: Optional[float]

The timeout for the Bedrock API request in seconds. It will be used for both connect and read timeouts.

type: Optional[Literal["BEDROCK_EMBEDDING"]]

Type of the embedding model.

class CohereEmbedding: …
api_key: Optional[str]

The Cohere API key.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embedding_type: Optional[str]

Embedding type. If not provided float embedding_type is used when needed.

input_type: Optional[str]

Model Input type. If not provided, search_document and search_query are used when needed.

model_name: Optional[str]

The modelId of the Cohere model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

truncate: Optional[str]

Truncation type - START/ END/ NONE

class CohereEmbeddingConfig: …
component: Optional[CohereEmbedding]

Configuration for the Cohere embedding model.

api_key: Optional[str]

The Cohere API key.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embedding_type: Optional[str]

Embedding type. If not provided float embedding_type is used when needed.

input_type: Optional[str]

Model Input type. If not provided, search_document and search_query are used when needed.

model_name: Optional[str]

The modelId of the Cohere model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

truncate: Optional[str]

Truncation type - START/ END/ NONE

type: Optional[Literal["COHERE_EMBEDDING"]]

Type of the embedding model.

class DataSinkCreate: …

Schema for creating a data sink.

component: Component

Component that implements the data sink

One of the following:
Dict[str, object]
class CloudPineconeVectorStore: …

Cloud Pinecone Vector Store.

This class is used to store the configuration for a Pinecone vector store, so that it can be created and used in LlamaCloud.

Args: api_key (str): API key for authenticating with Pinecone index_name (str): name of the Pinecone index namespace (optional[str]): namespace to use in the Pinecone index insert_kwargs (optional[dict]): additional kwargs to pass during insertion

api_key: str

The API key for authenticating with Pinecone

formatpassword
index_name: str
class_name: Optional[str]
insert_kwargs: Optional[Dict[str, object]]
namespace: Optional[str]
supports_nested_metadata_filters: Optional[Literal[true]]
class CloudPostgresVectorStore: …
database: str
embed_dim: int
host: str
password: str
port: int
schema_name: str
table_name: str
user: str
class_name: Optional[str]
hnsw_settings: Optional[PgVectorHnswSettings]

HNSW settings for PGVector.

distance_method: Optional[Literal["l2", "ip", "cosine", 3 more]]

The distance method to use.

One of the following:
"l2"
"ip"
"cosine"
"l1"
"hamming"
"jaccard"
ef_construction: Optional[int]

The number of edges to use during the construction phase.

minimum1

The number of edges to use during the search phase.

minimum1
m: Optional[int]

The number of bi-directional links created for each new element.

minimum1
vector_type: Optional[Literal["vector", "half_vec", "bit", "sparse_vec"]]

The type of vector to use.

One of the following:
"vector"
"half_vec"
"bit"
"sparse_vec"
perform_setup: Optional[bool]
supports_nested_metadata_filters: Optional[bool]
class CloudQdrantVectorStore: …

Cloud Qdrant Vector Store.

This class is used to store the configuration for a Qdrant vector store, so that it can be created and used in LlamaCloud.

Args: collection_name (str): name of the Qdrant collection url (str): url of the Qdrant instance api_key (str): API key for authenticating with Qdrant max_retries (int): maximum number of retries in case of a failure. Defaults to 3 client_kwargs (dict): additional kwargs to pass to the Qdrant client

api_key: str
collection_name: str
url: str
class_name: Optional[str]
client_kwargs: Optional[Dict[str, object]]
max_retries: Optional[int]
supports_nested_metadata_filters: Optional[Literal[true]]
class CloudAzureAISearchVectorStore: …

Cloud Azure AI Search Vector Store.

search_service_api_key: str
search_service_endpoint: str
class_name: Optional[str]
client_id: Optional[str]
client_secret: Optional[str]
embedding_dimension: Optional[int]
filterable_metadata_field_keys: Optional[Dict[str, object]]
index_name: Optional[str]
search_service_api_version: Optional[str]
supports_nested_metadata_filters: Optional[Literal[true]]
tenant_id: Optional[str]

Cloud MongoDB Atlas Vector Store.

This class is used to store the configuration for a MongoDB Atlas vector store, so that it can be created and used in LlamaCloud.

Args: mongodb_uri (str): URI for connecting to MongoDB Atlas db_name (str): name of the MongoDB database collection_name (str): name of the MongoDB collection vector_index_name (str): name of the MongoDB Atlas vector index fulltext_index_name (str): name of the MongoDB Atlas full-text index

class CloudMilvusVectorStore: …

Cloud Milvus Vector Store.

uri: str
token: Optional[str]
class_name: Optional[str]
collection_name: Optional[str]
embedding_dimension: Optional[int]
supports_nested_metadata_filters: Optional[bool]
class CloudAstraDBVectorStore: …

Cloud AstraDB Vector Store.

This class is used to store the configuration for an AstraDB vector store, so that it can be created and used in LlamaCloud.

Args: token (str): The Astra DB Application Token to use. api_endpoint (str): The Astra DB JSON API endpoint for your database. collection_name (str): Collection name to use. If not existing, it will be created. embedding_dimension (int): Length of the embedding vectors in use. keyspace (optional[str]): The keyspace to use. If not provided, ‘default_keyspace’

token: str

The Astra DB Application Token to use

formatpassword
api_endpoint: str

The Astra DB JSON API endpoint for your database

collection_name: str

Collection name to use. If not existing, it will be created

embedding_dimension: int

Length of the embedding vectors in use

class_name: Optional[str]
keyspace: Optional[str]

The keyspace to use. If not provided, ‘default_keyspace’

supports_nested_metadata_filters: Optional[Literal[true]]
name: str

The name of the data sink.

sink_type: Literal["PINECONE", "POSTGRES", "QDRANT", 4 more]
One of the following:
"PINECONE"
"POSTGRES"
"QDRANT"
"AZUREAI_SEARCH"
"MONGODB_ATLAS"
"MILVUS"
"ASTRA_DB"
class GeminiEmbedding: …
api_base: Optional[str]

API base to access the model. Defaults to None.

api_key: Optional[str]

API key to access the model. Defaults to None.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
model_name: Optional[str]

The modelId of the Gemini model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

output_dimensionality: Optional[int]

Optional reduced dimension for output embeddings. Supported by models/text-embedding-004 and newer (e.g. gemini-embedding-001). Not supported by models/embedding-001.

task_type: Optional[str]

The task for embedding model.

title: Optional[str]

Title is only applicable for retrieval_document tasks, and is used to represent a document title. For other tasks, title is invalid.

transport: Optional[str]

Transport to access the model. Defaults to None.

class GeminiEmbeddingConfig: …
component: Optional[GeminiEmbedding]

Configuration for the Gemini embedding model.

api_base: Optional[str]

API base to access the model. Defaults to None.

api_key: Optional[str]

API key to access the model. Defaults to None.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
model_name: Optional[str]

The modelId of the Gemini model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

output_dimensionality: Optional[int]

Optional reduced dimension for output embeddings. Supported by models/text-embedding-004 and newer (e.g. gemini-embedding-001). Not supported by models/embedding-001.

task_type: Optional[str]

The task for embedding model.

title: Optional[str]

Title is only applicable for retrieval_document tasks, and is used to represent a document title. For other tasks, title is invalid.

transport: Optional[str]

Transport to access the model. Defaults to None.

type: Optional[Literal["GEMINI_EMBEDDING"]]

Type of the embedding model.

class HuggingFaceInferenceAPIEmbedding: …
token: Optional[Union[str, bool, null]]

Hugging Face token. Will default to the locally saved token. Pass token=False if you don’t want to send your token to the server.

One of the following:
str
bool
class_name: Optional[str]
cookies: Optional[Dict[str, str]]

Additional cookies to send to the server.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
headers: Optional[Dict[str, str]]

Additional headers to send to the server. By default only the authorization and user-agent headers are sent. Values in this dictionary will override the default values.

model_name: Optional[str]

Hugging Face model name. If None, the task will be used.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

pooling: Optional[Literal["cls", "mean", "last"]]

Enum of possible pooling choices with pooling behaviors.

One of the following:
"cls"
"mean"
"last"
query_instruction: Optional[str]

Instruction to prepend during query embedding.

task: Optional[str]

Optional task to pick Hugging Face’s recommended model, used when model_name is left as default of None.

text_instruction: Optional[str]

Instruction to prepend during text embedding.

timeout: Optional[float]

The maximum number of seconds to wait for a response from the server. Loading a new model in Inference API can take up to several minutes. Defaults to None, meaning it will loop until the server is available.

class HuggingFaceInferenceAPIEmbeddingConfig: …
component: Optional[HuggingFaceInferenceAPIEmbedding]

Configuration for the HuggingFace Inference API embedding model.

token: Optional[Union[str, bool, null]]

Hugging Face token. Will default to the locally saved token. Pass token=False if you don’t want to send your token to the server.

One of the following:
str
bool
class_name: Optional[str]
cookies: Optional[Dict[str, str]]

Additional cookies to send to the server.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
headers: Optional[Dict[str, str]]

Additional headers to send to the server. By default only the authorization and user-agent headers are sent. Values in this dictionary will override the default values.

model_name: Optional[str]

Hugging Face model name. If None, the task will be used.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

pooling: Optional[Literal["cls", "mean", "last"]]

Enum of possible pooling choices with pooling behaviors.

One of the following:
"cls"
"mean"
"last"
query_instruction: Optional[str]

Instruction to prepend during query embedding.

task: Optional[str]

Optional task to pick Hugging Face’s recommended model, used when model_name is left as default of None.

text_instruction: Optional[str]

Instruction to prepend during text embedding.

timeout: Optional[float]

The maximum number of seconds to wait for a response from the server. Loading a new model in Inference API can take up to several minutes. Defaults to None, meaning it will loop until the server is available.

type: Optional[Literal["HUGGINGFACE_API_EMBEDDING"]]

Type of the embedding model.

class LlamaParseParameters: …

Settings that can be configured for how to use LlamaParse to parse files within a LlamaCloud pipeline.

adaptive_long_table: Optional[bool]
aggressive_table_extraction: Optional[bool]
auto_mode: Optional[bool]
auto_mode_configuration_json: Optional[str]
auto_mode_trigger_on_image_in_page: Optional[bool]
auto_mode_trigger_on_regexp_in_page: Optional[str]
auto_mode_trigger_on_table_in_page: Optional[bool]
auto_mode_trigger_on_text_in_page: Optional[str]
azure_openai_api_version: Optional[str]
azure_openai_deployment_name: Optional[str]
azure_openai_endpoint: Optional[str]
azure_openai_key: Optional[str]
bbox_bottom: Optional[float]
bbox_left: Optional[float]
bbox_right: Optional[float]
bbox_top: Optional[float]
bounding_box: Optional[str]
compact_markdown_table: Optional[bool]
complemental_formatting_instruction: Optional[str]
content_guideline_instruction: Optional[str]
continuous_mode: Optional[bool]
disable_image_extraction: Optional[bool]
disable_ocr: Optional[bool]
disable_reconstruction: Optional[bool]
do_not_cache: Optional[bool]
do_not_unroll_columns: Optional[bool]
enable_cost_optimizer: Optional[bool]
extract_charts: Optional[bool]
extract_layout: Optional[bool]
extract_printed_page_number: Optional[bool]
fast_mode: Optional[bool]
formatting_instruction: Optional[str]
gpt4o_api_key: Optional[str]
gpt4o_mode: Optional[bool]
guess_xlsx_sheet_name: Optional[bool]
hide_footers: Optional[bool]
hide_headers: Optional[bool]
high_res_ocr: Optional[bool]
html_make_all_elements_visible: Optional[bool]
html_remove_fixed_elements: Optional[bool]
html_remove_navigation_elements: Optional[bool]
http_proxy: Optional[str]
ignore_document_elements_for_layout_detection: Optional[bool]
images_to_save: Optional[List[Literal["screenshot", "embedded", "layout"]]]
One of the following:
"screenshot"
"embedded"
"layout"
inline_images_in_markdown: Optional[bool]
input_s3_path: Optional[str]
input_s3_region: Optional[str]
input_url: Optional[str]
internal_is_screenshot_job: Optional[bool]
invalidate_cache: Optional[bool]
is_formatting_instruction: Optional[bool]
job_timeout_extra_time_per_page_in_seconds: Optional[float]
job_timeout_in_seconds: Optional[float]
keep_page_separator_when_merging_tables: Optional[bool]
languages: Optional[List[ParsingLanguages]]
One of the following:
"af"
"az"
"bs"
"cs"
"cy"
"da"
"de"
"en"
"es"
"et"
"fr"
"ga"
"hr"
"hu"
"id"
"is"
"it"
"ku"
"la"
"lt"
"lv"
"mi"
"ms"
"mt"
"nl"
"no"
"oc"
"pi"
"pl"
"pt"
"ro"
"rs_latin"
"sk"
"sl"
"sq"
"sv"
"sw"
"tl"
"tr"
"uz"
"vi"
"ar"
"fa"
"ug"
"ur"
"bn"
"as"
"mni"
"ru"
"rs_cyrillic"
"be"
"bg"
"uk"
"mn"
"abq"
"ady"
"kbd"
"ava"
"dar"
"inh"
"che"
"lbe"
"lez"
"tab"
"tjk"
"hi"
"mr"
"ne"
"bh"
"mai"
"ang"
"bho"
"mah"
"sck"
"new"
"gom"
"sa"
"bgc"
"th"
"ch_sim"
"ch_tra"
"ja"
"ko"
"ta"
"te"
"kn"
layout_aware: Optional[bool]
line_level_bounding_box: Optional[bool]
markdown_table_multiline_header_separator: Optional[str]
max_pages: Optional[int]
max_pages_enforced: Optional[int]
merge_tables_across_pages_in_markdown: Optional[bool]
model: Optional[str]
outlined_table_extraction: Optional[bool]
output_pdf_of_document: Optional[bool]
output_s3_path_prefix: Optional[str]
output_s3_region: Optional[str]
output_tables_as_html: Optional[bool]
page_error_tolerance: Optional[float]
page_footer_prefix: Optional[str]
page_footer_suffix: Optional[str]
page_header_prefix: Optional[str]
page_header_suffix: Optional[str]
page_prefix: Optional[str]
page_separator: Optional[str]
page_suffix: Optional[str]
parse_mode: Optional[ParsingMode]

Enum for representing the mode of parsing to be used.

One of the following:
"parse_page_without_llm"
"parse_page_with_llm"
"parse_page_with_lvm"
"parse_page_with_agent"
"parse_page_with_layout_agent"
"parse_document_with_llm"
"parse_document_with_lvm"
"parse_document_with_agent"
parsing_instruction: Optional[str]
precise_bounding_box: Optional[bool]
premium_mode: Optional[bool]
presentation_out_of_bounds_content: Optional[bool]
presentation_skip_embedded_data: Optional[bool]
preserve_layout_alignment_across_pages: Optional[bool]
preserve_very_small_text: Optional[bool]
preset: Optional[str]
priority: Optional[Literal["low", "medium", "high", "critical"]]

The priority for the request. This field may be ignored or overwritten depending on the organization tier.

One of the following:
"low"
"medium"
"high"
"critical"
project_id: Optional[str]
remove_hidden_text: Optional[bool]
replace_failed_page_mode: Optional[FailPageMode]

Enum for representing the different available page error handling modes.

One of the following:
"raw_text"
"blank_page"
"error_message"
replace_failed_page_with_error_message_prefix: Optional[str]
replace_failed_page_with_error_message_suffix: Optional[str]
save_images: Optional[bool]
skip_diagonal_text: Optional[bool]
specialized_chart_parsing_agentic: Optional[bool]
specialized_chart_parsing_efficient: Optional[bool]
specialized_chart_parsing_plus: Optional[bool]
specialized_image_parsing: Optional[bool]
spreadsheet_extract_sub_tables: Optional[bool]
spreadsheet_force_formula_computation: Optional[bool]
spreadsheet_include_hidden_sheets: Optional[bool]
strict_mode_buggy_font: Optional[bool]
strict_mode_image_extraction: Optional[bool]
strict_mode_image_ocr: Optional[bool]
strict_mode_reconstruction: Optional[bool]
structured_output: Optional[bool]
structured_output_json_schema: Optional[str]
structured_output_json_schema_name: Optional[str]
system_prompt: Optional[str]
system_prompt_append: Optional[str]
take_screenshot: Optional[bool]
target_pages: Optional[str]
tier: Optional[str]
use_vendor_multimodal_model: Optional[bool]
user_prompt: Optional[str]
vendor_multimodal_api_key: Optional[str]
vendor_multimodal_model_name: Optional[str]
version: Optional[str]
webhook_configurations: Optional[List[WebhookConfiguration]]

Outbound webhook endpoints to notify on job status changes

webhook_events: Optional[List[Literal["extract.pending", "extract.success", "extract.error", 14 more]]]

Events to subscribe to (e.g. ‘parse.success’, ‘extract.error’). If null, all events are delivered.

One of the following:
"extract.pending"
"extract.success"
"extract.error"
"extract.partial_success"
"extract.cancelled"
"parse.pending"
"parse.running"
"parse.success"
"parse.error"
"parse.partial_success"
"parse.cancelled"
"classify.pending"
"classify.success"
"classify.error"
"classify.partial_success"
"classify.cancelled"
"unmapped_event"
webhook_headers: Optional[Dict[str, str]]

Custom HTTP headers sent with each webhook request (e.g. auth tokens)

webhook_output_format: Optional[str]

Response format sent to the webhook: ‘string’ (default) or ‘json’

webhook_url: Optional[str]

URL to receive webhook POST notifications

webhook_url: Optional[str]
class LlmParameters: …
class_name: Optional[str]
model_name: Optional[Literal["GPT_4O", "GPT_4O_MINI", "GPT_4_1", 10 more]]

The name of the model to use for LLM completions.

One of the following:
"GPT_4O"
"GPT_4O_MINI"
"GPT_4_1"
"GPT_4_1_NANO"
"GPT_4_1_MINI"
"AZURE_OPENAI_GPT_4O"
"AZURE_OPENAI_GPT_4O_MINI"
"AZURE_OPENAI_GPT_4_1"
"AZURE_OPENAI_GPT_4_1_MINI"
"AZURE_OPENAI_GPT_4_1_NANO"
"CLAUDE_4_5_SONNET"
"BEDROCK_CLAUDE_3_5_SONNET_V1"
"BEDROCK_CLAUDE_3_5_SONNET_V2"
system_prompt: Optional[str]

The system prompt to use for the completion.

maxLength3000
temperature: Optional[float]

The temperature value for the model.

use_chain_of_thought_reasoning: Optional[bool]

Whether to use chain of thought reasoning.

use_citation: Optional[bool]

Whether to show citations in the response.

class ManagedIngestionStatusResponse: …
status: Literal["NOT_STARTED", "IN_PROGRESS", "SUCCESS", 3 more]

Status of the ingestion.

One of the following:
"NOT_STARTED"
"IN_PROGRESS"
"SUCCESS"
"ERROR"
"PARTIAL_SUCCESS"
"CANCELLED"
deployment_date: Optional[datetime]

Date of the deployment.

formatdate-time
effective_at: Optional[datetime]

When the status is effective

formatdate-time
error: Optional[List[Error]]

List of errors that occurred during ingestion.

job_id: str

ID of the job that failed.

formatuuid
message: str

List of errors that occurred during ingestion.

step: Literal["MANAGED_INGESTION", "DATA_SOURCE", "FILE_UPDATER", 4 more]

Name of the job that failed.

One of the following:
"MANAGED_INGESTION"
"DATA_SOURCE"
"FILE_UPDATER"
"PARSE"
"TRANSFORM"
"INGESTION"
"METADATA_UPDATE"
job_id: Optional[str]

ID of the latest job.

formatuuid
Literal["system", "developer", "user", 5 more]

Message role.

One of the following:
"system"
"developer"
"user"
"assistant"
"function"
"tool"
"chatbot"
"model"
class MetadataFilters: …

Metadata filters for vector stores.

filters: List[Filter]
One of the following:
class FilterMetadataFilter: …

Comprehensive metadata filter for vector stores to support more operators.

Value uses Strict types, as int, float and str are compatible types and were all converted to string before.

See: https://docs.pydantic.dev/latest/usage/types/#strict-types

key: str
value: Union[float, str, List[str], 3 more]
One of the following:
float
str
List[str]
List[float]
List[int]
operator: Optional[Literal["==", ">", "<", 11 more]]

Vector store filter operator.

One of the following:
"=="
">"
"<"
"!="
">="
"<="
"in"
"nin"
"any"
"all"
"text_match"
"text_match_insensitive"
"contains"
"is_empty"
condition: Optional[Literal["and", "or", "not"]]

Vector store filter conditions to combine different filters.

One of the following:
"and"
"or"
"not"
class OpenAIEmbedding: …
additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for OpenAI API.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for OpenAI API.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
class OpenAIEmbeddingConfig: …
component: Optional[OpenAIEmbedding]

Configuration for the OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for OpenAI API.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for OpenAI API.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["OPENAI_EMBEDDING"]]

Type of the embedding model.

class PageFigureNodeWithScore: …

Page figure metadata with score

node: Node
confidence: float

The confidence of the figure

maximum1
minimum0
figure_name: str

The name of the figure

figure_size: int

The size of the figure in bytes

minimum0
file_id: str

The ID of the file that the figure was taken from

formatuuid
page_index: int

The index of the page for which the figure is taken (0-indexed)

minimum0
is_likely_noise: Optional[bool]

Whether the figure is likely to be noise

metadata: Optional[Dict[str, object]]

Metadata for the figure

score: float

The score of the figure node

class_name: Optional[str]
class PageScreenshotNodeWithScore: …

Page screenshot metadata with score

node: Node
file_id: str

The ID of the file that the page screenshot was taken from

formatuuid
image_size: int

The size of the image in bytes

minimum0
page_index: int

The index of the page for which the screenshot is taken (0-indexed)

minimum0
metadata: Optional[Dict[str, object]]

Metadata for the screenshot

score: float

The score of the screenshot node

class_name: Optional[str]
class Pipeline: …

Schema for a pipeline.

id: str

Unique identifier

formatuuid
embedding_config: EmbeddingConfig
One of the following:
class EmbeddingConfigManagedOpenAIEmbeddingConfig: …
component: Optional[EmbeddingConfigManagedOpenAIEmbeddingConfigComponent]

Configuration for the Managed OpenAI embedding model.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
model_name: Optional[Literal["openai-text-embedding-3-small"]]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

type: Optional[Literal["MANAGED_OPENAI_EMBEDDING"]]

Type of the embedding model.

class AzureOpenAIEmbeddingConfig: …
component: Optional[AzureOpenAIEmbedding]

Configuration for the Azure OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for Azure deployment.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for Azure OpenAI API.

azure_deployment: Optional[str]

The Azure deployment to use.

azure_endpoint: Optional[str]

The Azure endpoint to use.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["AZURE_EMBEDDING"]]

Type of the embedding model.

class CohereEmbeddingConfig: …
component: Optional[CohereEmbedding]

Configuration for the Cohere embedding model.

api_key: Optional[str]

The Cohere API key.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embedding_type: Optional[str]

Embedding type. If not provided float embedding_type is used when needed.

input_type: Optional[str]

Model Input type. If not provided, search_document and search_query are used when needed.

model_name: Optional[str]

The modelId of the Cohere model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

truncate: Optional[str]

Truncation type - START/ END/ NONE

type: Optional[Literal["COHERE_EMBEDDING"]]

Type of the embedding model.

class GeminiEmbeddingConfig: …
component: Optional[GeminiEmbedding]

Configuration for the Gemini embedding model.

api_base: Optional[str]

API base to access the model. Defaults to None.

api_key: Optional[str]

API key to access the model. Defaults to None.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
model_name: Optional[str]

The modelId of the Gemini model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

output_dimensionality: Optional[int]

Optional reduced dimension for output embeddings. Supported by models/text-embedding-004 and newer (e.g. gemini-embedding-001). Not supported by models/embedding-001.

task_type: Optional[str]

The task for embedding model.

title: Optional[str]

Title is only applicable for retrieval_document tasks, and is used to represent a document title. For other tasks, title is invalid.

transport: Optional[str]

Transport to access the model. Defaults to None.

type: Optional[Literal["GEMINI_EMBEDDING"]]

Type of the embedding model.

class HuggingFaceInferenceAPIEmbeddingConfig: …
component: Optional[HuggingFaceInferenceAPIEmbedding]

Configuration for the HuggingFace Inference API embedding model.

token: Optional[Union[str, bool, null]]

Hugging Face token. Will default to the locally saved token. Pass token=False if you don’t want to send your token to the server.

One of the following:
str
bool
class_name: Optional[str]
cookies: Optional[Dict[str, str]]

Additional cookies to send to the server.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
headers: Optional[Dict[str, str]]

Additional headers to send to the server. By default only the authorization and user-agent headers are sent. Values in this dictionary will override the default values.

model_name: Optional[str]

Hugging Face model name. If None, the task will be used.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

pooling: Optional[Literal["cls", "mean", "last"]]

Enum of possible pooling choices with pooling behaviors.

One of the following:
"cls"
"mean"
"last"
query_instruction: Optional[str]

Instruction to prepend during query embedding.

task: Optional[str]

Optional task to pick Hugging Face’s recommended model, used when model_name is left as default of None.

text_instruction: Optional[str]

Instruction to prepend during text embedding.

timeout: Optional[float]

The maximum number of seconds to wait for a response from the server. Loading a new model in Inference API can take up to several minutes. Defaults to None, meaning it will loop until the server is available.

type: Optional[Literal["HUGGINGFACE_API_EMBEDDING"]]

Type of the embedding model.

class OpenAIEmbeddingConfig: …
component: Optional[OpenAIEmbedding]

Configuration for the OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for OpenAI API.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for OpenAI API.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["OPENAI_EMBEDDING"]]

Type of the embedding model.

class VertexAIEmbeddingConfig: …
component: Optional[VertexTextEmbedding]

Configuration for the VertexAI embedding model.

client_email: Optional[str]

The client email for the VertexAI credentials.

location: str

The default location to use when making API calls.

private_key: Optional[str]

The private key for the VertexAI credentials.

private_key_id: Optional[str]

The private key ID for the VertexAI credentials.

project: str

The default GCP project to use when making Vertex API calls.

token_uri: Optional[str]

The token URI for the VertexAI credentials.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the Vertex.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embed_mode: Optional[Literal["default", "classification", "clustering", 2 more]]

The embedding mode to use.

One of the following:
"default"
"classification"
"clustering"
"similarity"
"retrieval"
model_name: Optional[str]

The modelId of the VertexAI model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

type: Optional[Literal["VERTEXAI_EMBEDDING"]]

Type of the embedding model.

class BedrockEmbeddingConfig: …
component: Optional[BedrockEmbedding]

Configuration for the Bedrock embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the bedrock client.

aws_access_key_id: Optional[str]

AWS Access Key ID to use

aws_secret_access_key: Optional[str]

AWS Secret Access Key to use

aws_session_token: Optional[str]

AWS Session Token to use

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

The maximum number of API retries.

exclusiveMinimum0
model_name: Optional[str]

The modelId of the Bedrock model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

profile_name: Optional[str]

The name of aws profile to use. If not given, then the default profile is used.

region_name: Optional[str]

AWS region name to use. Uses region configured in AWS CLI if not passed

timeout: Optional[float]

The timeout for the Bedrock API request in seconds. It will be used for both connect and read timeouts.

type: Optional[Literal["BEDROCK_EMBEDDING"]]

Type of the embedding model.

name: str
project_id: str
config_hash: Optional[ConfigHash]

Hashes for the configuration of a pipeline.

embedding_config_hash: Optional[str]

Hash of the embedding config.

parsing_config_hash: Optional[str]

Hash of the llama parse parameters.

transform_config_hash: Optional[str]

Hash of the transform config.

created_at: Optional[datetime]

Creation datetime

formatdate-time
data_sink: Optional[DataSink]

Schema for a data sink.

id: str

Unique identifier

formatuuid
component: Component

Component that implements the data sink

One of the following:
Dict[str, object]
class CloudPineconeVectorStore: …

Cloud Pinecone Vector Store.

This class is used to store the configuration for a Pinecone vector store, so that it can be created and used in LlamaCloud.

Args: api_key (str): API key for authenticating with Pinecone index_name (str): name of the Pinecone index namespace (optional[str]): namespace to use in the Pinecone index insert_kwargs (optional[dict]): additional kwargs to pass during insertion

api_key: str

The API key for authenticating with Pinecone

formatpassword
index_name: str
class_name: Optional[str]
insert_kwargs: Optional[Dict[str, object]]
namespace: Optional[str]
supports_nested_metadata_filters: Optional[Literal[true]]
class CloudPostgresVectorStore: …
database: str
embed_dim: int
host: str
password: str
port: int
schema_name: str
table_name: str
user: str
class_name: Optional[str]
hnsw_settings: Optional[PgVectorHnswSettings]

HNSW settings for PGVector.

distance_method: Optional[Literal["l2", "ip", "cosine", 3 more]]

The distance method to use.

One of the following:
"l2"
"ip"
"cosine"
"l1"
"hamming"
"jaccard"
ef_construction: Optional[int]

The number of edges to use during the construction phase.

minimum1

The number of edges to use during the search phase.

minimum1
m: Optional[int]

The number of bi-directional links created for each new element.

minimum1
vector_type: Optional[Literal["vector", "half_vec", "bit", "sparse_vec"]]

The type of vector to use.

One of the following:
"vector"
"half_vec"
"bit"
"sparse_vec"
perform_setup: Optional[bool]
supports_nested_metadata_filters: Optional[bool]
class CloudQdrantVectorStore: …

Cloud Qdrant Vector Store.

This class is used to store the configuration for a Qdrant vector store, so that it can be created and used in LlamaCloud.

Args: collection_name (str): name of the Qdrant collection url (str): url of the Qdrant instance api_key (str): API key for authenticating with Qdrant max_retries (int): maximum number of retries in case of a failure. Defaults to 3 client_kwargs (dict): additional kwargs to pass to the Qdrant client

api_key: str
collection_name: str
url: str
class_name: Optional[str]
client_kwargs: Optional[Dict[str, object]]
max_retries: Optional[int]
supports_nested_metadata_filters: Optional[Literal[true]]
class CloudAzureAISearchVectorStore: …

Cloud Azure AI Search Vector Store.

search_service_api_key: str
search_service_endpoint: str
class_name: Optional[str]
client_id: Optional[str]
client_secret: Optional[str]
embedding_dimension: Optional[int]
filterable_metadata_field_keys: Optional[Dict[str, object]]
index_name: Optional[str]
search_service_api_version: Optional[str]
supports_nested_metadata_filters: Optional[Literal[true]]
tenant_id: Optional[str]

Cloud MongoDB Atlas Vector Store.

This class is used to store the configuration for a MongoDB Atlas vector store, so that it can be created and used in LlamaCloud.

Args: mongodb_uri (str): URI for connecting to MongoDB Atlas db_name (str): name of the MongoDB database collection_name (str): name of the MongoDB collection vector_index_name (str): name of the MongoDB Atlas vector index fulltext_index_name (str): name of the MongoDB Atlas full-text index

class CloudMilvusVectorStore: …

Cloud Milvus Vector Store.

uri: str
token: Optional[str]
class_name: Optional[str]
collection_name: Optional[str]
embedding_dimension: Optional[int]
supports_nested_metadata_filters: Optional[bool]
class CloudAstraDBVectorStore: …

Cloud AstraDB Vector Store.

This class is used to store the configuration for an AstraDB vector store, so that it can be created and used in LlamaCloud.

Args: token (str): The Astra DB Application Token to use. api_endpoint (str): The Astra DB JSON API endpoint for your database. collection_name (str): Collection name to use. If not existing, it will be created. embedding_dimension (int): Length of the embedding vectors in use. keyspace (optional[str]): The keyspace to use. If not provided, ‘default_keyspace’

token: str

The Astra DB Application Token to use

formatpassword
api_endpoint: str

The Astra DB JSON API endpoint for your database

collection_name: str

Collection name to use. If not existing, it will be created

embedding_dimension: int

Length of the embedding vectors in use

class_name: Optional[str]
keyspace: Optional[str]

The keyspace to use. If not provided, ‘default_keyspace’

supports_nested_metadata_filters: Optional[Literal[true]]
name: str

The name of the data sink.

project_id: str
sink_type: Literal["PINECONE", "POSTGRES", "QDRANT", 4 more]
One of the following:
"PINECONE"
"POSTGRES"
"QDRANT"
"AZUREAI_SEARCH"
"MONGODB_ATLAS"
"MILVUS"
"ASTRA_DB"
created_at: Optional[datetime]

Creation datetime

formatdate-time
updated_at: Optional[datetime]

Update datetime

formatdate-time
embedding_model_config: Optional[EmbeddingModelConfig]

Schema for an embedding model config.

id: str

Unique identifier

formatuuid
embedding_config: EmbeddingModelConfigEmbeddingConfig

The embedding configuration for the embedding model config.

One of the following:
class AzureOpenAIEmbeddingConfig: …
component: Optional[AzureOpenAIEmbedding]

Configuration for the Azure OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for Azure deployment.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for Azure OpenAI API.

azure_deployment: Optional[str]

The Azure deployment to use.

azure_endpoint: Optional[str]

The Azure endpoint to use.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["AZURE_EMBEDDING"]]

Type of the embedding model.

class CohereEmbeddingConfig: …
component: Optional[CohereEmbedding]

Configuration for the Cohere embedding model.

api_key: Optional[str]

The Cohere API key.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embedding_type: Optional[str]

Embedding type. If not provided float embedding_type is used when needed.

input_type: Optional[str]

Model Input type. If not provided, search_document and search_query are used when needed.

model_name: Optional[str]

The modelId of the Cohere model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

truncate: Optional[str]

Truncation type - START/ END/ NONE

type: Optional[Literal["COHERE_EMBEDDING"]]

Type of the embedding model.

class GeminiEmbeddingConfig: …
component: Optional[GeminiEmbedding]

Configuration for the Gemini embedding model.

api_base: Optional[str]

API base to access the model. Defaults to None.

api_key: Optional[str]

API key to access the model. Defaults to None.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
model_name: Optional[str]

The modelId of the Gemini model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

output_dimensionality: Optional[int]

Optional reduced dimension for output embeddings. Supported by models/text-embedding-004 and newer (e.g. gemini-embedding-001). Not supported by models/embedding-001.

task_type: Optional[str]

The task for embedding model.

title: Optional[str]

Title is only applicable for retrieval_document tasks, and is used to represent a document title. For other tasks, title is invalid.

transport: Optional[str]

Transport to access the model. Defaults to None.

type: Optional[Literal["GEMINI_EMBEDDING"]]

Type of the embedding model.

class HuggingFaceInferenceAPIEmbeddingConfig: …
component: Optional[HuggingFaceInferenceAPIEmbedding]

Configuration for the HuggingFace Inference API embedding model.

token: Optional[Union[str, bool, null]]

Hugging Face token. Will default to the locally saved token. Pass token=False if you don’t want to send your token to the server.

One of the following:
str
bool
class_name: Optional[str]
cookies: Optional[Dict[str, str]]

Additional cookies to send to the server.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
headers: Optional[Dict[str, str]]

Additional headers to send to the server. By default only the authorization and user-agent headers are sent. Values in this dictionary will override the default values.

model_name: Optional[str]

Hugging Face model name. If None, the task will be used.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

pooling: Optional[Literal["cls", "mean", "last"]]

Enum of possible pooling choices with pooling behaviors.

One of the following:
"cls"
"mean"
"last"
query_instruction: Optional[str]

Instruction to prepend during query embedding.

task: Optional[str]

Optional task to pick Hugging Face’s recommended model, used when model_name is left as default of None.

text_instruction: Optional[str]

Instruction to prepend during text embedding.

timeout: Optional[float]

The maximum number of seconds to wait for a response from the server. Loading a new model in Inference API can take up to several minutes. Defaults to None, meaning it will loop until the server is available.

type: Optional[Literal["HUGGINGFACE_API_EMBEDDING"]]

Type of the embedding model.

class OpenAIEmbeddingConfig: …
component: Optional[OpenAIEmbedding]

Configuration for the OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for OpenAI API.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for OpenAI API.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["OPENAI_EMBEDDING"]]

Type of the embedding model.

class VertexAIEmbeddingConfig: …
component: Optional[VertexTextEmbedding]

Configuration for the VertexAI embedding model.

client_email: Optional[str]

The client email for the VertexAI credentials.

location: str

The default location to use when making API calls.

private_key: Optional[str]

The private key for the VertexAI credentials.

private_key_id: Optional[str]

The private key ID for the VertexAI credentials.

project: str

The default GCP project to use when making Vertex API calls.

token_uri: Optional[str]

The token URI for the VertexAI credentials.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the Vertex.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embed_mode: Optional[Literal["default", "classification", "clustering", 2 more]]

The embedding mode to use.

One of the following:
"default"
"classification"
"clustering"
"similarity"
"retrieval"
model_name: Optional[str]

The modelId of the VertexAI model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

type: Optional[Literal["VERTEXAI_EMBEDDING"]]

Type of the embedding model.

class BedrockEmbeddingConfig: …
component: Optional[BedrockEmbedding]

Configuration for the Bedrock embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the bedrock client.

aws_access_key_id: Optional[str]

AWS Access Key ID to use

aws_secret_access_key: Optional[str]

AWS Secret Access Key to use

aws_session_token: Optional[str]

AWS Session Token to use

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

The maximum number of API retries.

exclusiveMinimum0
model_name: Optional[str]

The modelId of the Bedrock model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

profile_name: Optional[str]

The name of aws profile to use. If not given, then the default profile is used.

region_name: Optional[str]

AWS region name to use. Uses region configured in AWS CLI if not passed

timeout: Optional[float]

The timeout for the Bedrock API request in seconds. It will be used for both connect and read timeouts.

type: Optional[Literal["BEDROCK_EMBEDDING"]]

Type of the embedding model.

name: str

The name of the embedding model config.

project_id: str
created_at: Optional[datetime]

Creation datetime

formatdate-time
updated_at: Optional[datetime]

Update datetime

formatdate-time
embedding_model_config_id: Optional[str]

The ID of the EmbeddingModelConfig this pipeline is using.

formatuuid
llama_parse_parameters: Optional[LlamaParseParameters]

Settings that can be configured for how to use LlamaParse to parse files within a LlamaCloud pipeline.

adaptive_long_table: Optional[bool]
aggressive_table_extraction: Optional[bool]
auto_mode: Optional[bool]
auto_mode_configuration_json: Optional[str]
auto_mode_trigger_on_image_in_page: Optional[bool]
auto_mode_trigger_on_regexp_in_page: Optional[str]
auto_mode_trigger_on_table_in_page: Optional[bool]
auto_mode_trigger_on_text_in_page: Optional[str]
azure_openai_api_version: Optional[str]
azure_openai_deployment_name: Optional[str]
azure_openai_endpoint: Optional[str]
azure_openai_key: Optional[str]
bbox_bottom: Optional[float]
bbox_left: Optional[float]
bbox_right: Optional[float]
bbox_top: Optional[float]
bounding_box: Optional[str]
compact_markdown_table: Optional[bool]
complemental_formatting_instruction: Optional[str]
content_guideline_instruction: Optional[str]
continuous_mode: Optional[bool]
disable_image_extraction: Optional[bool]
disable_ocr: Optional[bool]
disable_reconstruction: Optional[bool]
do_not_cache: Optional[bool]
do_not_unroll_columns: Optional[bool]
enable_cost_optimizer: Optional[bool]
extract_charts: Optional[bool]
extract_layout: Optional[bool]
extract_printed_page_number: Optional[bool]
fast_mode: Optional[bool]
formatting_instruction: Optional[str]
gpt4o_api_key: Optional[str]
gpt4o_mode: Optional[bool]
guess_xlsx_sheet_name: Optional[bool]
hide_footers: Optional[bool]
hide_headers: Optional[bool]
high_res_ocr: Optional[bool]
html_make_all_elements_visible: Optional[bool]
html_remove_fixed_elements: Optional[bool]
html_remove_navigation_elements: Optional[bool]
http_proxy: Optional[str]
ignore_document_elements_for_layout_detection: Optional[bool]
images_to_save: Optional[List[Literal["screenshot", "embedded", "layout"]]]
One of the following:
"screenshot"
"embedded"
"layout"
inline_images_in_markdown: Optional[bool]
input_s3_path: Optional[str]
input_s3_region: Optional[str]
input_url: Optional[str]
internal_is_screenshot_job: Optional[bool]
invalidate_cache: Optional[bool]
is_formatting_instruction: Optional[bool]
job_timeout_extra_time_per_page_in_seconds: Optional[float]
job_timeout_in_seconds: Optional[float]
keep_page_separator_when_merging_tables: Optional[bool]
languages: Optional[List[ParsingLanguages]]
One of the following:
"af"
"az"
"bs"
"cs"
"cy"
"da"
"de"
"en"
"es"
"et"
"fr"
"ga"
"hr"
"hu"
"id"
"is"
"it"
"ku"
"la"
"lt"
"lv"
"mi"
"ms"
"mt"
"nl"
"no"
"oc"
"pi"
"pl"
"pt"
"ro"
"rs_latin"
"sk"
"sl"
"sq"
"sv"
"sw"
"tl"
"tr"
"uz"
"vi"
"ar"
"fa"
"ug"
"ur"
"bn"
"as"
"mni"
"ru"
"rs_cyrillic"
"be"
"bg"
"uk"
"mn"
"abq"
"ady"
"kbd"
"ava"
"dar"
"inh"
"che"
"lbe"
"lez"
"tab"
"tjk"
"hi"
"mr"
"ne"
"bh"
"mai"
"ang"
"bho"
"mah"
"sck"
"new"
"gom"
"sa"
"bgc"
"th"
"ch_sim"
"ch_tra"
"ja"
"ko"
"ta"
"te"
"kn"
layout_aware: Optional[bool]
line_level_bounding_box: Optional[bool]
markdown_table_multiline_header_separator: Optional[str]
max_pages: Optional[int]
max_pages_enforced: Optional[int]
merge_tables_across_pages_in_markdown: Optional[bool]
model: Optional[str]
outlined_table_extraction: Optional[bool]
output_pdf_of_document: Optional[bool]
output_s3_path_prefix: Optional[str]
output_s3_region: Optional[str]
output_tables_as_html: Optional[bool]
page_error_tolerance: Optional[float]
page_footer_prefix: Optional[str]
page_footer_suffix: Optional[str]
page_header_prefix: Optional[str]
page_header_suffix: Optional[str]
page_prefix: Optional[str]
page_separator: Optional[str]
page_suffix: Optional[str]
parse_mode: Optional[ParsingMode]

Enum for representing the mode of parsing to be used.

One of the following:
"parse_page_without_llm"
"parse_page_with_llm"
"parse_page_with_lvm"
"parse_page_with_agent"
"parse_page_with_layout_agent"
"parse_document_with_llm"
"parse_document_with_lvm"
"parse_document_with_agent"
parsing_instruction: Optional[str]
precise_bounding_box: Optional[bool]
premium_mode: Optional[bool]
presentation_out_of_bounds_content: Optional[bool]
presentation_skip_embedded_data: Optional[bool]
preserve_layout_alignment_across_pages: Optional[bool]
preserve_very_small_text: Optional[bool]
preset: Optional[str]
priority: Optional[Literal["low", "medium", "high", "critical"]]

The priority for the request. This field may be ignored or overwritten depending on the organization tier.

One of the following:
"low"
"medium"
"high"
"critical"
project_id: Optional[str]
remove_hidden_text: Optional[bool]
replace_failed_page_mode: Optional[FailPageMode]

Enum for representing the different available page error handling modes.

One of the following:
"raw_text"
"blank_page"
"error_message"
replace_failed_page_with_error_message_prefix: Optional[str]
replace_failed_page_with_error_message_suffix: Optional[str]
save_images: Optional[bool]
skip_diagonal_text: Optional[bool]
specialized_chart_parsing_agentic: Optional[bool]
specialized_chart_parsing_efficient: Optional[bool]
specialized_chart_parsing_plus: Optional[bool]
specialized_image_parsing: Optional[bool]
spreadsheet_extract_sub_tables: Optional[bool]
spreadsheet_force_formula_computation: Optional[bool]
spreadsheet_include_hidden_sheets: Optional[bool]
strict_mode_buggy_font: Optional[bool]
strict_mode_image_extraction: Optional[bool]
strict_mode_image_ocr: Optional[bool]
strict_mode_reconstruction: Optional[bool]
structured_output: Optional[bool]
structured_output_json_schema: Optional[str]
structured_output_json_schema_name: Optional[str]
system_prompt: Optional[str]
system_prompt_append: Optional[str]
take_screenshot: Optional[bool]
target_pages: Optional[str]
tier: Optional[str]
use_vendor_multimodal_model: Optional[bool]
user_prompt: Optional[str]
vendor_multimodal_api_key: Optional[str]
vendor_multimodal_model_name: Optional[str]
version: Optional[str]
webhook_configurations: Optional[List[WebhookConfiguration]]

Outbound webhook endpoints to notify on job status changes

webhook_events: Optional[List[Literal["extract.pending", "extract.success", "extract.error", 14 more]]]

Events to subscribe to (e.g. ‘parse.success’, ‘extract.error’). If null, all events are delivered.

One of the following:
"extract.pending"
"extract.success"
"extract.error"
"extract.partial_success"
"extract.cancelled"
"parse.pending"
"parse.running"
"parse.success"
"parse.error"
"parse.partial_success"
"parse.cancelled"
"classify.pending"
"classify.success"
"classify.error"
"classify.partial_success"
"classify.cancelled"
"unmapped_event"
webhook_headers: Optional[Dict[str, str]]

Custom HTTP headers sent with each webhook request (e.g. auth tokens)

webhook_output_format: Optional[str]

Response format sent to the webhook: ‘string’ (default) or ‘json’

webhook_url: Optional[str]

URL to receive webhook POST notifications

webhook_url: Optional[str]
managed_pipeline_id: Optional[str]

The ID of the ManagedPipeline this playground pipeline is linked to.

formatuuid
metadata_config: Optional[PipelineMetadataConfig]

Metadata configuration for the pipeline.

excluded_embed_metadata_keys: Optional[List[str]]

List of metadata keys to exclude from embeddings

excluded_llm_metadata_keys: Optional[List[str]]

List of metadata keys to exclude from LLM during retrieval

pipeline_type: Optional[PipelineType]

Type of pipeline. Either PLAYGROUND or MANAGED.

One of the following:
"PLAYGROUND"
"MANAGED"
preset_retrieval_parameters: Optional[PresetRetrievalParams]

Preset retrieval parameters for the pipeline.

alpha: Optional[float]

Alpha value for hybrid retrieval to determine the weights between dense and sparse retrieval. 0 is sparse retrieval and 1 is dense retrieval.

maximum1
minimum0
class_name: Optional[str]
dense_similarity_cutoff: Optional[float]

Minimum similarity score wrt query for retrieval

maximum1
minimum0
dense_similarity_top_k: Optional[int]

Number of nodes for dense retrieval.

maximum100
minimum1
enable_reranking: Optional[bool]

Enable reranking for retrieval

files_top_k: Optional[int]

Number of files to retrieve (only for retrieval mode files_via_metadata and files_via_content).

maximum5
minimum1
rerank_top_n: Optional[int]

Number of reranked nodes for returning.

maximum100
minimum1
retrieval_mode: Optional[RetrievalMode]

The retrieval mode for the query.

One of the following:
"chunks"
"files_via_metadata"
"files_via_content"
"auto_routed"
Deprecatedretrieve_image_nodes: Optional[bool]

Whether to retrieve image nodes.

retrieve_page_figure_nodes: Optional[bool]

Whether to retrieve page figure nodes.

retrieve_page_screenshot_nodes: Optional[bool]

Whether to retrieve page screenshot nodes.

search_filters: Optional[MetadataFilters]

Metadata filters for vector stores.

filters: List[Filter]
One of the following:
class FilterMetadataFilter: …

Comprehensive metadata filter for vector stores to support more operators.

Value uses Strict types, as int, float and str are compatible types and were all converted to string before.

See: https://docs.pydantic.dev/latest/usage/types/#strict-types

key: str
value: Union[float, str, List[str], 3 more]
One of the following:
float
str
List[str]
List[float]
List[int]
operator: Optional[Literal["==", ">", "<", 11 more]]

Vector store filter operator.

One of the following:
"=="
">"
"<"
"!="
">="
"<="
"in"
"nin"
"any"
"all"
"text_match"
"text_match_insensitive"
"contains"
"is_empty"
condition: Optional[Literal["and", "or", "not"]]

Vector store filter conditions to combine different filters.

One of the following:
"and"
"or"
"not"
search_filters_inference_schema: Optional[Dict[str, Union[Dict[str, object], List[object], str, 3 more]]]

JSON Schema that will be used to infer search_filters. Omit or leave as null to skip inference.

One of the following:
Dict[str, object]
List[object]
str
float
bool
sparse_similarity_top_k: Optional[int]

Number of nodes for sparse retrieval.

maximum100
minimum1
sparse_model_config: Optional[SparseModelConfig]

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.

class_name: Optional[str]
model_type: Optional[Literal["splade", "bm25", "auto"]]

The sparse model type to use. ‘bm25’ uses Qdrant’s FastEmbed BM25 model (default for new pipelines), ‘splade’ uses HuggingFace Splade model, ‘auto’ selects based on deployment mode (BYOC uses term frequency, Cloud uses Splade).

One of the following:
"splade"
"bm25"
"auto"
status: Optional[Literal["CREATED", "DELETING"]]

Status of the pipeline.

One of the following:
"CREATED"
"DELETING"
transform_config: Optional[TransformConfig]

Configuration for the transformation.

One of the following:
class AutoTransformConfig: …
chunk_overlap: Optional[int]

Chunk overlap for the transformation.

chunk_size: Optional[int]

Chunk size for the transformation.

exclusiveMinimum0
mode: Optional[Literal["auto"]]
class AdvancedModeTransformConfig: …
chunking_config: Optional[ChunkingConfig]

Configuration for the chunking.

One of the following:
class ChunkingConfigNoneChunkingConfig: …
mode: Optional[Literal["none"]]
class ChunkingConfigCharacterChunkingConfig: …
chunk_overlap: Optional[int]
chunk_size: Optional[int]
mode: Optional[Literal["character"]]
class ChunkingConfigTokenChunkingConfig: …
chunk_overlap: Optional[int]
chunk_size: Optional[int]
mode: Optional[Literal["token"]]
separator: Optional[str]
class ChunkingConfigSentenceChunkingConfig: …
chunk_overlap: Optional[int]
chunk_size: Optional[int]
mode: Optional[Literal["sentence"]]
paragraph_separator: Optional[str]
separator: Optional[str]
class ChunkingConfigSemanticChunkingConfig: …
breakpoint_percentile_threshold: Optional[int]
buffer_size: Optional[int]
mode: Optional[Literal["semantic"]]
mode: Optional[Literal["advanced"]]
segmentation_config: Optional[SegmentationConfig]

Configuration for the segmentation.

One of the following:
class SegmentationConfigNoneSegmentationConfig: …
mode: Optional[Literal["none"]]
class SegmentationConfigPageSegmentationConfig: …
mode: Optional[Literal["page"]]
page_separator: Optional[str]
class SegmentationConfigElementSegmentationConfig: …
mode: Optional[Literal["element"]]
updated_at: Optional[datetime]

Update datetime

formatdate-time
class PipelineCreate: …

Schema for creating a pipeline.

name: str
data_sink: Optional[DataSinkCreate]

Schema for creating a data sink.

component: Component

Component that implements the data sink

One of the following:
Dict[str, object]
class CloudPineconeVectorStore: …

Cloud Pinecone Vector Store.

This class is used to store the configuration for a Pinecone vector store, so that it can be created and used in LlamaCloud.

Args: api_key (str): API key for authenticating with Pinecone index_name (str): name of the Pinecone index namespace (optional[str]): namespace to use in the Pinecone index insert_kwargs (optional[dict]): additional kwargs to pass during insertion

api_key: str

The API key for authenticating with Pinecone

formatpassword
index_name: str
class_name: Optional[str]
insert_kwargs: Optional[Dict[str, object]]
namespace: Optional[str]
supports_nested_metadata_filters: Optional[Literal[true]]
class CloudPostgresVectorStore: …
database: str
embed_dim: int
host: str
password: str
port: int
schema_name: str
table_name: str
user: str
class_name: Optional[str]
hnsw_settings: Optional[PgVectorHnswSettings]

HNSW settings for PGVector.

distance_method: Optional[Literal["l2", "ip", "cosine", 3 more]]

The distance method to use.

One of the following:
"l2"
"ip"
"cosine"
"l1"
"hamming"
"jaccard"
ef_construction: Optional[int]

The number of edges to use during the construction phase.

minimum1

The number of edges to use during the search phase.

minimum1
m: Optional[int]

The number of bi-directional links created for each new element.

minimum1
vector_type: Optional[Literal["vector", "half_vec", "bit", "sparse_vec"]]

The type of vector to use.

One of the following:
"vector"
"half_vec"
"bit"
"sparse_vec"
perform_setup: Optional[bool]
supports_nested_metadata_filters: Optional[bool]
class CloudQdrantVectorStore: …

Cloud Qdrant Vector Store.

This class is used to store the configuration for a Qdrant vector store, so that it can be created and used in LlamaCloud.

Args: collection_name (str): name of the Qdrant collection url (str): url of the Qdrant instance api_key (str): API key for authenticating with Qdrant max_retries (int): maximum number of retries in case of a failure. Defaults to 3 client_kwargs (dict): additional kwargs to pass to the Qdrant client

api_key: str
collection_name: str
url: str
class_name: Optional[str]
client_kwargs: Optional[Dict[str, object]]
max_retries: Optional[int]
supports_nested_metadata_filters: Optional[Literal[true]]
class CloudAzureAISearchVectorStore: …

Cloud Azure AI Search Vector Store.

search_service_api_key: str
search_service_endpoint: str
class_name: Optional[str]
client_id: Optional[str]
client_secret: Optional[str]
embedding_dimension: Optional[int]
filterable_metadata_field_keys: Optional[Dict[str, object]]
index_name: Optional[str]
search_service_api_version: Optional[str]
supports_nested_metadata_filters: Optional[Literal[true]]
tenant_id: Optional[str]

Cloud MongoDB Atlas Vector Store.

This class is used to store the configuration for a MongoDB Atlas vector store, so that it can be created and used in LlamaCloud.

Args: mongodb_uri (str): URI for connecting to MongoDB Atlas db_name (str): name of the MongoDB database collection_name (str): name of the MongoDB collection vector_index_name (str): name of the MongoDB Atlas vector index fulltext_index_name (str): name of the MongoDB Atlas full-text index

class CloudMilvusVectorStore: …

Cloud Milvus Vector Store.

uri: str
token: Optional[str]
class_name: Optional[str]
collection_name: Optional[str]
embedding_dimension: Optional[int]
supports_nested_metadata_filters: Optional[bool]
class CloudAstraDBVectorStore: …

Cloud AstraDB Vector Store.

This class is used to store the configuration for an AstraDB vector store, so that it can be created and used in LlamaCloud.

Args: token (str): The Astra DB Application Token to use. api_endpoint (str): The Astra DB JSON API endpoint for your database. collection_name (str): Collection name to use. If not existing, it will be created. embedding_dimension (int): Length of the embedding vectors in use. keyspace (optional[str]): The keyspace to use. If not provided, ‘default_keyspace’

token: str

The Astra DB Application Token to use

formatpassword
api_endpoint: str

The Astra DB JSON API endpoint for your database

collection_name: str

Collection name to use. If not existing, it will be created

embedding_dimension: int

Length of the embedding vectors in use

class_name: Optional[str]
keyspace: Optional[str]

The keyspace to use. If not provided, ‘default_keyspace’

supports_nested_metadata_filters: Optional[Literal[true]]
name: str

The name of the data sink.

sink_type: Literal["PINECONE", "POSTGRES", "QDRANT", 4 more]
One of the following:
"PINECONE"
"POSTGRES"
"QDRANT"
"AZUREAI_SEARCH"
"MONGODB_ATLAS"
"MILVUS"
"ASTRA_DB"
data_sink_id: Optional[str]

Data sink ID. When provided instead of data_sink, the data sink will be looked up by ID.

formatuuid
embedding_config: Optional[EmbeddingConfig]
One of the following:
class AzureOpenAIEmbeddingConfig: …
component: Optional[AzureOpenAIEmbedding]

Configuration for the Azure OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for Azure deployment.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for Azure OpenAI API.

azure_deployment: Optional[str]

The Azure deployment to use.

azure_endpoint: Optional[str]

The Azure endpoint to use.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["AZURE_EMBEDDING"]]

Type of the embedding model.

class CohereEmbeddingConfig: …
component: Optional[CohereEmbedding]

Configuration for the Cohere embedding model.

api_key: Optional[str]

The Cohere API key.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embedding_type: Optional[str]

Embedding type. If not provided float embedding_type is used when needed.

input_type: Optional[str]

Model Input type. If not provided, search_document and search_query are used when needed.

model_name: Optional[str]

The modelId of the Cohere model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

truncate: Optional[str]

Truncation type - START/ END/ NONE

type: Optional[Literal["COHERE_EMBEDDING"]]

Type of the embedding model.

class GeminiEmbeddingConfig: …
component: Optional[GeminiEmbedding]

Configuration for the Gemini embedding model.

api_base: Optional[str]

API base to access the model. Defaults to None.

api_key: Optional[str]

API key to access the model. Defaults to None.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
model_name: Optional[str]

The modelId of the Gemini model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

output_dimensionality: Optional[int]

Optional reduced dimension for output embeddings. Supported by models/text-embedding-004 and newer (e.g. gemini-embedding-001). Not supported by models/embedding-001.

task_type: Optional[str]

The task for embedding model.

title: Optional[str]

Title is only applicable for retrieval_document tasks, and is used to represent a document title. For other tasks, title is invalid.

transport: Optional[str]

Transport to access the model. Defaults to None.

type: Optional[Literal["GEMINI_EMBEDDING"]]

Type of the embedding model.

class HuggingFaceInferenceAPIEmbeddingConfig: …
component: Optional[HuggingFaceInferenceAPIEmbedding]

Configuration for the HuggingFace Inference API embedding model.

token: Optional[Union[str, bool, null]]

Hugging Face token. Will default to the locally saved token. Pass token=False if you don’t want to send your token to the server.

One of the following:
str
bool
class_name: Optional[str]
cookies: Optional[Dict[str, str]]

Additional cookies to send to the server.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
headers: Optional[Dict[str, str]]

Additional headers to send to the server. By default only the authorization and user-agent headers are sent. Values in this dictionary will override the default values.

model_name: Optional[str]

Hugging Face model name. If None, the task will be used.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

pooling: Optional[Literal["cls", "mean", "last"]]

Enum of possible pooling choices with pooling behaviors.

One of the following:
"cls"
"mean"
"last"
query_instruction: Optional[str]

Instruction to prepend during query embedding.

task: Optional[str]

Optional task to pick Hugging Face’s recommended model, used when model_name is left as default of None.

text_instruction: Optional[str]

Instruction to prepend during text embedding.

timeout: Optional[float]

The maximum number of seconds to wait for a response from the server. Loading a new model in Inference API can take up to several minutes. Defaults to None, meaning it will loop until the server is available.

type: Optional[Literal["HUGGINGFACE_API_EMBEDDING"]]

Type of the embedding model.

class OpenAIEmbeddingConfig: …
component: Optional[OpenAIEmbedding]

Configuration for the OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for OpenAI API.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for OpenAI API.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["OPENAI_EMBEDDING"]]

Type of the embedding model.

class VertexAIEmbeddingConfig: …
component: Optional[VertexTextEmbedding]

Configuration for the VertexAI embedding model.

client_email: Optional[str]

The client email for the VertexAI credentials.

location: str

The default location to use when making API calls.

private_key: Optional[str]

The private key for the VertexAI credentials.

private_key_id: Optional[str]

The private key ID for the VertexAI credentials.

project: str

The default GCP project to use when making Vertex API calls.

token_uri: Optional[str]

The token URI for the VertexAI credentials.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the Vertex.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embed_mode: Optional[Literal["default", "classification", "clustering", 2 more]]

The embedding mode to use.

One of the following:
"default"
"classification"
"clustering"
"similarity"
"retrieval"
model_name: Optional[str]

The modelId of the VertexAI model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

type: Optional[Literal["VERTEXAI_EMBEDDING"]]

Type of the embedding model.

class BedrockEmbeddingConfig: …
component: Optional[BedrockEmbedding]

Configuration for the Bedrock embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the bedrock client.

aws_access_key_id: Optional[str]

AWS Access Key ID to use

aws_secret_access_key: Optional[str]

AWS Secret Access Key to use

aws_session_token: Optional[str]

AWS Session Token to use

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

The maximum number of API retries.

exclusiveMinimum0
model_name: Optional[str]

The modelId of the Bedrock model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

profile_name: Optional[str]

The name of aws profile to use. If not given, then the default profile is used.

region_name: Optional[str]

AWS region name to use. Uses region configured in AWS CLI if not passed

timeout: Optional[float]

The timeout for the Bedrock API request in seconds. It will be used for both connect and read timeouts.

type: Optional[Literal["BEDROCK_EMBEDDING"]]

Type of the embedding model.

embedding_model_config_id: Optional[str]

Embedding model config ID. When provided instead of embedding_config, the embedding model config will be looked up by ID.

formatuuid
llama_parse_parameters: Optional[LlamaParseParameters]

Settings that can be configured for how to use LlamaParse to parse files within a LlamaCloud pipeline.

adaptive_long_table: Optional[bool]
aggressive_table_extraction: Optional[bool]
auto_mode: Optional[bool]
auto_mode_configuration_json: Optional[str]
auto_mode_trigger_on_image_in_page: Optional[bool]
auto_mode_trigger_on_regexp_in_page: Optional[str]
auto_mode_trigger_on_table_in_page: Optional[bool]
auto_mode_trigger_on_text_in_page: Optional[str]
azure_openai_api_version: Optional[str]
azure_openai_deployment_name: Optional[str]
azure_openai_endpoint: Optional[str]
azure_openai_key: Optional[str]
bbox_bottom: Optional[float]
bbox_left: Optional[float]
bbox_right: Optional[float]
bbox_top: Optional[float]
bounding_box: Optional[str]
compact_markdown_table: Optional[bool]
complemental_formatting_instruction: Optional[str]
content_guideline_instruction: Optional[str]
continuous_mode: Optional[bool]
disable_image_extraction: Optional[bool]
disable_ocr: Optional[bool]
disable_reconstruction: Optional[bool]
do_not_cache: Optional[bool]
do_not_unroll_columns: Optional[bool]
enable_cost_optimizer: Optional[bool]
extract_charts: Optional[bool]
extract_layout: Optional[bool]
extract_printed_page_number: Optional[bool]
fast_mode: Optional[bool]
formatting_instruction: Optional[str]
gpt4o_api_key: Optional[str]
gpt4o_mode: Optional[bool]
guess_xlsx_sheet_name: Optional[bool]
hide_footers: Optional[bool]
hide_headers: Optional[bool]
high_res_ocr: Optional[bool]
html_make_all_elements_visible: Optional[bool]
html_remove_fixed_elements: Optional[bool]
html_remove_navigation_elements: Optional[bool]
http_proxy: Optional[str]
ignore_document_elements_for_layout_detection: Optional[bool]
images_to_save: Optional[List[Literal["screenshot", "embedded", "layout"]]]
One of the following:
"screenshot"
"embedded"
"layout"
inline_images_in_markdown: Optional[bool]
input_s3_path: Optional[str]
input_s3_region: Optional[str]
input_url: Optional[str]
internal_is_screenshot_job: Optional[bool]
invalidate_cache: Optional[bool]
is_formatting_instruction: Optional[bool]
job_timeout_extra_time_per_page_in_seconds: Optional[float]
job_timeout_in_seconds: Optional[float]
keep_page_separator_when_merging_tables: Optional[bool]
languages: Optional[List[ParsingLanguages]]
One of the following:
"af"
"az"
"bs"
"cs"
"cy"
"da"
"de"
"en"
"es"
"et"
"fr"
"ga"
"hr"
"hu"
"id"
"is"
"it"
"ku"
"la"
"lt"
"lv"
"mi"
"ms"
"mt"
"nl"
"no"
"oc"
"pi"
"pl"
"pt"
"ro"
"rs_latin"
"sk"
"sl"
"sq"
"sv"
"sw"
"tl"
"tr"
"uz"
"vi"
"ar"
"fa"
"ug"
"ur"
"bn"
"as"
"mni"
"ru"
"rs_cyrillic"
"be"
"bg"
"uk"
"mn"
"abq"
"ady"
"kbd"
"ava"
"dar"
"inh"
"che"
"lbe"
"lez"
"tab"
"tjk"
"hi"
"mr"
"ne"
"bh"
"mai"
"ang"
"bho"
"mah"
"sck"
"new"
"gom"
"sa"
"bgc"
"th"
"ch_sim"
"ch_tra"
"ja"
"ko"
"ta"
"te"
"kn"
layout_aware: Optional[bool]
line_level_bounding_box: Optional[bool]
markdown_table_multiline_header_separator: Optional[str]
max_pages: Optional[int]
max_pages_enforced: Optional[int]
merge_tables_across_pages_in_markdown: Optional[bool]
model: Optional[str]
outlined_table_extraction: Optional[bool]
output_pdf_of_document: Optional[bool]
output_s3_path_prefix: Optional[str]
output_s3_region: Optional[str]
output_tables_as_html: Optional[bool]
page_error_tolerance: Optional[float]
page_footer_prefix: Optional[str]
page_footer_suffix: Optional[str]
page_header_prefix: Optional[str]
page_header_suffix: Optional[str]
page_prefix: Optional[str]
page_separator: Optional[str]
page_suffix: Optional[str]
parse_mode: Optional[ParsingMode]

Enum for representing the mode of parsing to be used.

One of the following:
"parse_page_without_llm"
"parse_page_with_llm"
"parse_page_with_lvm"
"parse_page_with_agent"
"parse_page_with_layout_agent"
"parse_document_with_llm"
"parse_document_with_lvm"
"parse_document_with_agent"
parsing_instruction: Optional[str]
precise_bounding_box: Optional[bool]
premium_mode: Optional[bool]
presentation_out_of_bounds_content: Optional[bool]
presentation_skip_embedded_data: Optional[bool]
preserve_layout_alignment_across_pages: Optional[bool]
preserve_very_small_text: Optional[bool]
preset: Optional[str]
priority: Optional[Literal["low", "medium", "high", "critical"]]

The priority for the request. This field may be ignored or overwritten depending on the organization tier.

One of the following:
"low"
"medium"
"high"
"critical"
project_id: Optional[str]
remove_hidden_text: Optional[bool]
replace_failed_page_mode: Optional[FailPageMode]

Enum for representing the different available page error handling modes.

One of the following:
"raw_text"
"blank_page"
"error_message"
replace_failed_page_with_error_message_prefix: Optional[str]
replace_failed_page_with_error_message_suffix: Optional[str]
save_images: Optional[bool]
skip_diagonal_text: Optional[bool]
specialized_chart_parsing_agentic: Optional[bool]
specialized_chart_parsing_efficient: Optional[bool]
specialized_chart_parsing_plus: Optional[bool]
specialized_image_parsing: Optional[bool]
spreadsheet_extract_sub_tables: Optional[bool]
spreadsheet_force_formula_computation: Optional[bool]
spreadsheet_include_hidden_sheets: Optional[bool]
strict_mode_buggy_font: Optional[bool]
strict_mode_image_extraction: Optional[bool]
strict_mode_image_ocr: Optional[bool]
strict_mode_reconstruction: Optional[bool]
structured_output: Optional[bool]
structured_output_json_schema: Optional[str]
structured_output_json_schema_name: Optional[str]
system_prompt: Optional[str]
system_prompt_append: Optional[str]
take_screenshot: Optional[bool]
target_pages: Optional[str]
tier: Optional[str]
use_vendor_multimodal_model: Optional[bool]
user_prompt: Optional[str]
vendor_multimodal_api_key: Optional[str]
vendor_multimodal_model_name: Optional[str]
version: Optional[str]
webhook_configurations: Optional[List[WebhookConfiguration]]

Outbound webhook endpoints to notify on job status changes

webhook_events: Optional[List[Literal["extract.pending", "extract.success", "extract.error", 14 more]]]

Events to subscribe to (e.g. ‘parse.success’, ‘extract.error’). If null, all events are delivered.

One of the following:
"extract.pending"
"extract.success"
"extract.error"
"extract.partial_success"
"extract.cancelled"
"parse.pending"
"parse.running"
"parse.success"
"parse.error"
"parse.partial_success"
"parse.cancelled"
"classify.pending"
"classify.success"
"classify.error"
"classify.partial_success"
"classify.cancelled"
"unmapped_event"
webhook_headers: Optional[Dict[str, str]]

Custom HTTP headers sent with each webhook request (e.g. auth tokens)

webhook_output_format: Optional[str]

Response format sent to the webhook: ‘string’ (default) or ‘json’

webhook_url: Optional[str]

URL to receive webhook POST notifications

webhook_url: Optional[str]
managed_pipeline_id: Optional[str]

The ID of the ManagedPipeline this playground pipeline is linked to.

formatuuid
metadata_config: Optional[PipelineMetadataConfig]

Metadata configuration for the pipeline.

excluded_embed_metadata_keys: Optional[List[str]]

List of metadata keys to exclude from embeddings

excluded_llm_metadata_keys: Optional[List[str]]

List of metadata keys to exclude from LLM during retrieval

pipeline_type: Optional[PipelineType]

Type of pipeline. Either PLAYGROUND or MANAGED.

One of the following:
"PLAYGROUND"
"MANAGED"
preset_retrieval_parameters: Optional[PresetRetrievalParams]

Preset retrieval parameters for the pipeline.

alpha: Optional[float]

Alpha value for hybrid retrieval to determine the weights between dense and sparse retrieval. 0 is sparse retrieval and 1 is dense retrieval.

maximum1
minimum0
class_name: Optional[str]
dense_similarity_cutoff: Optional[float]

Minimum similarity score wrt query for retrieval

maximum1
minimum0
dense_similarity_top_k: Optional[int]

Number of nodes for dense retrieval.

maximum100
minimum1
enable_reranking: Optional[bool]

Enable reranking for retrieval

files_top_k: Optional[int]

Number of files to retrieve (only for retrieval mode files_via_metadata and files_via_content).

maximum5
minimum1
rerank_top_n: Optional[int]

Number of reranked nodes for returning.

maximum100
minimum1
retrieval_mode: Optional[RetrievalMode]

The retrieval mode for the query.

One of the following:
"chunks"
"files_via_metadata"
"files_via_content"
"auto_routed"
Deprecatedretrieve_image_nodes: Optional[bool]

Whether to retrieve image nodes.

retrieve_page_figure_nodes: Optional[bool]

Whether to retrieve page figure nodes.

retrieve_page_screenshot_nodes: Optional[bool]

Whether to retrieve page screenshot nodes.

search_filters: Optional[MetadataFilters]

Metadata filters for vector stores.

filters: List[Filter]
One of the following:
class FilterMetadataFilter: …

Comprehensive metadata filter for vector stores to support more operators.

Value uses Strict types, as int, float and str are compatible types and were all converted to string before.

See: https://docs.pydantic.dev/latest/usage/types/#strict-types

key: str
value: Union[float, str, List[str], 3 more]
One of the following:
float
str
List[str]
List[float]
List[int]
operator: Optional[Literal["==", ">", "<", 11 more]]

Vector store filter operator.

One of the following:
"=="
">"
"<"
"!="
">="
"<="
"in"
"nin"
"any"
"all"
"text_match"
"text_match_insensitive"
"contains"
"is_empty"
condition: Optional[Literal["and", "or", "not"]]

Vector store filter conditions to combine different filters.

One of the following:
"and"
"or"
"not"
search_filters_inference_schema: Optional[Dict[str, Union[Dict[str, object], List[object], str, 3 more]]]

JSON Schema that will be used to infer search_filters. Omit or leave as null to skip inference.

One of the following:
Dict[str, object]
List[object]
str
float
bool
sparse_similarity_top_k: Optional[int]

Number of nodes for sparse retrieval.

maximum100
minimum1
sparse_model_config: Optional[SparseModelConfig]

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.

class_name: Optional[str]
model_type: Optional[Literal["splade", "bm25", "auto"]]

The sparse model type to use. ‘bm25’ uses Qdrant’s FastEmbed BM25 model (default for new pipelines), ‘splade’ uses HuggingFace Splade model, ‘auto’ selects based on deployment mode (BYOC uses term frequency, Cloud uses Splade).

One of the following:
"splade"
"bm25"
"auto"
status: Optional[str]

Status of the pipeline deployment.

transform_config: Optional[TransformConfig]

Configuration for the transformation.

One of the following:
class AutoTransformConfig: …
chunk_overlap: Optional[int]

Chunk overlap for the transformation.

chunk_size: Optional[int]

Chunk size for the transformation.

exclusiveMinimum0
mode: Optional[Literal["auto"]]
class AdvancedModeTransformConfig: …
chunking_config: Optional[ChunkingConfig]

Configuration for the chunking.

One of the following:
class ChunkingConfigNoneChunkingConfig: …
mode: Optional[Literal["none"]]
class ChunkingConfigCharacterChunkingConfig: …
chunk_overlap: Optional[int]
chunk_size: Optional[int]
mode: Optional[Literal["character"]]
class ChunkingConfigTokenChunkingConfig: …
chunk_overlap: Optional[int]
chunk_size: Optional[int]
mode: Optional[Literal["token"]]
separator: Optional[str]
class ChunkingConfigSentenceChunkingConfig: …
chunk_overlap: Optional[int]
chunk_size: Optional[int]
mode: Optional[Literal["sentence"]]
paragraph_separator: Optional[str]
separator: Optional[str]
class ChunkingConfigSemanticChunkingConfig: …
breakpoint_percentile_threshold: Optional[int]
buffer_size: Optional[int]
mode: Optional[Literal["semantic"]]
mode: Optional[Literal["advanced"]]
segmentation_config: Optional[SegmentationConfig]

Configuration for the segmentation.

One of the following:
class SegmentationConfigNoneSegmentationConfig: …
mode: Optional[Literal["none"]]
class SegmentationConfigPageSegmentationConfig: …
mode: Optional[Literal["page"]]
page_separator: Optional[str]
class SegmentationConfigElementSegmentationConfig: …
mode: Optional[Literal["element"]]
class PipelineMetadataConfig: …
excluded_embed_metadata_keys: Optional[List[str]]

List of metadata keys to exclude from embeddings

excluded_llm_metadata_keys: Optional[List[str]]

List of metadata keys to exclude from LLM during retrieval

Literal["PLAYGROUND", "MANAGED"]

Enum for representing the type of a pipeline

One of the following:
"PLAYGROUND"
"MANAGED"
class PresetRetrievalParams: …

Schema for the search params for an retrieval execution that can be preset for a pipeline.

alpha: Optional[float]

Alpha value for hybrid retrieval to determine the weights between dense and sparse retrieval. 0 is sparse retrieval and 1 is dense retrieval.

maximum1
minimum0
class_name: Optional[str]
dense_similarity_cutoff: Optional[float]

Minimum similarity score wrt query for retrieval

maximum1
minimum0
dense_similarity_top_k: Optional[int]

Number of nodes for dense retrieval.

maximum100
minimum1
enable_reranking: Optional[bool]

Enable reranking for retrieval

files_top_k: Optional[int]

Number of files to retrieve (only for retrieval mode files_via_metadata and files_via_content).

maximum5
minimum1
rerank_top_n: Optional[int]

Number of reranked nodes for returning.

maximum100
minimum1
retrieval_mode: Optional[RetrievalMode]

The retrieval mode for the query.

One of the following:
"chunks"
"files_via_metadata"
"files_via_content"
"auto_routed"
Deprecatedretrieve_image_nodes: Optional[bool]

Whether to retrieve image nodes.

retrieve_page_figure_nodes: Optional[bool]

Whether to retrieve page figure nodes.

retrieve_page_screenshot_nodes: Optional[bool]

Whether to retrieve page screenshot nodes.

search_filters: Optional[MetadataFilters]

Metadata filters for vector stores.

filters: List[Filter]
One of the following:
class FilterMetadataFilter: …

Comprehensive metadata filter for vector stores to support more operators.

Value uses Strict types, as int, float and str are compatible types and were all converted to string before.

See: https://docs.pydantic.dev/latest/usage/types/#strict-types

key: str
value: Union[float, str, List[str], 3 more]
One of the following:
float
str
List[str]
List[float]
List[int]
operator: Optional[Literal["==", ">", "<", 11 more]]

Vector store filter operator.

One of the following:
"=="
">"
"<"
"!="
">="
"<="
"in"
"nin"
"any"
"all"
"text_match"
"text_match_insensitive"
"contains"
"is_empty"
condition: Optional[Literal["and", "or", "not"]]

Vector store filter conditions to combine different filters.

One of the following:
"and"
"or"
"not"
search_filters_inference_schema: Optional[Dict[str, Union[Dict[str, object], List[object], str, 3 more]]]

JSON Schema that will be used to infer search_filters. Omit or leave as null to skip inference.

One of the following:
Dict[str, object]
List[object]
str
float
bool
sparse_similarity_top_k: Optional[int]

Number of nodes for sparse retrieval.

maximum100
minimum1
Literal["chunks", "files_via_metadata", "files_via_content", "auto_routed"]
One of the following:
"chunks"
"files_via_metadata"
"files_via_content"
"auto_routed"
class SparseModelConfig: …

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.

class_name: Optional[str]
model_type: Optional[Literal["splade", "bm25", "auto"]]

The sparse model type to use. ‘bm25’ uses Qdrant’s FastEmbed BM25 model (default for new pipelines), ‘splade’ uses HuggingFace Splade model, ‘auto’ selects based on deployment mode (BYOC uses term frequency, Cloud uses Splade).

One of the following:
"splade"
"bm25"
"auto"
class VertexAIEmbeddingConfig: …
component: Optional[VertexTextEmbedding]

Configuration for the VertexAI embedding model.

client_email: Optional[str]

The client email for the VertexAI credentials.

location: str

The default location to use when making API calls.

private_key: Optional[str]

The private key for the VertexAI credentials.

private_key_id: Optional[str]

The private key ID for the VertexAI credentials.

project: str

The default GCP project to use when making Vertex API calls.

token_uri: Optional[str]

The token URI for the VertexAI credentials.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the Vertex.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embed_mode: Optional[Literal["default", "classification", "clustering", 2 more]]

The embedding mode to use.

One of the following:
"default"
"classification"
"clustering"
"similarity"
"retrieval"
model_name: Optional[str]

The modelId of the VertexAI model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

type: Optional[Literal["VERTEXAI_EMBEDDING"]]

Type of the embedding model.

class VertexTextEmbedding: …
client_email: Optional[str]

The client email for the VertexAI credentials.

location: str

The default location to use when making API calls.

private_key: Optional[str]

The private key for the VertexAI credentials.

private_key_id: Optional[str]

The private key ID for the VertexAI credentials.

project: str

The default GCP project to use when making Vertex API calls.

token_uri: Optional[str]

The token URI for the VertexAI credentials.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the Vertex.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embed_mode: Optional[Literal["default", "classification", "clustering", 2 more]]

The embedding mode to use.

One of the following:
"default"
"classification"
"clustering"
"similarity"
"retrieval"
model_name: Optional[str]

The modelId of the VertexAI model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

List[Pipeline]
id: str

Unique identifier

formatuuid
embedding_config: EmbeddingConfig
One of the following:
class EmbeddingConfigManagedOpenAIEmbeddingConfig: …
component: Optional[EmbeddingConfigManagedOpenAIEmbeddingConfigComponent]

Configuration for the Managed OpenAI embedding model.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
model_name: Optional[Literal["openai-text-embedding-3-small"]]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

type: Optional[Literal["MANAGED_OPENAI_EMBEDDING"]]

Type of the embedding model.

class AzureOpenAIEmbeddingConfig: …
component: Optional[AzureOpenAIEmbedding]

Configuration for the Azure OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for Azure deployment.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for Azure OpenAI API.

azure_deployment: Optional[str]

The Azure deployment to use.

azure_endpoint: Optional[str]

The Azure endpoint to use.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["AZURE_EMBEDDING"]]

Type of the embedding model.

class CohereEmbeddingConfig: …
component: Optional[CohereEmbedding]

Configuration for the Cohere embedding model.

api_key: Optional[str]

The Cohere API key.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embedding_type: Optional[str]

Embedding type. If not provided float embedding_type is used when needed.

input_type: Optional[str]

Model Input type. If not provided, search_document and search_query are used when needed.

model_name: Optional[str]

The modelId of the Cohere model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

truncate: Optional[str]

Truncation type - START/ END/ NONE

type: Optional[Literal["COHERE_EMBEDDING"]]

Type of the embedding model.

class GeminiEmbeddingConfig: …
component: Optional[GeminiEmbedding]

Configuration for the Gemini embedding model.

api_base: Optional[str]

API base to access the model. Defaults to None.

api_key: Optional[str]

API key to access the model. Defaults to None.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
model_name: Optional[str]

The modelId of the Gemini model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

output_dimensionality: Optional[int]

Optional reduced dimension for output embeddings. Supported by models/text-embedding-004 and newer (e.g. gemini-embedding-001). Not supported by models/embedding-001.

task_type: Optional[str]

The task for embedding model.

title: Optional[str]

Title is only applicable for retrieval_document tasks, and is used to represent a document title. For other tasks, title is invalid.

transport: Optional[str]

Transport to access the model. Defaults to None.

type: Optional[Literal["GEMINI_EMBEDDING"]]

Type of the embedding model.

class HuggingFaceInferenceAPIEmbeddingConfig: …
component: Optional[HuggingFaceInferenceAPIEmbedding]

Configuration for the HuggingFace Inference API embedding model.

token: Optional[Union[str, bool, null]]

Hugging Face token. Will default to the locally saved token. Pass token=False if you don’t want to send your token to the server.

One of the following:
str
bool
class_name: Optional[str]
cookies: Optional[Dict[str, str]]

Additional cookies to send to the server.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
headers: Optional[Dict[str, str]]

Additional headers to send to the server. By default only the authorization and user-agent headers are sent. Values in this dictionary will override the default values.

model_name: Optional[str]

Hugging Face model name. If None, the task will be used.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

pooling: Optional[Literal["cls", "mean", "last"]]

Enum of possible pooling choices with pooling behaviors.

One of the following:
"cls"
"mean"
"last"
query_instruction: Optional[str]

Instruction to prepend during query embedding.

task: Optional[str]

Optional task to pick Hugging Face’s recommended model, used when model_name is left as default of None.

text_instruction: Optional[str]

Instruction to prepend during text embedding.

timeout: Optional[float]

The maximum number of seconds to wait for a response from the server. Loading a new model in Inference API can take up to several minutes. Defaults to None, meaning it will loop until the server is available.

type: Optional[Literal["HUGGINGFACE_API_EMBEDDING"]]

Type of the embedding model.

class OpenAIEmbeddingConfig: …
component: Optional[OpenAIEmbedding]

Configuration for the OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for OpenAI API.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for OpenAI API.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["OPENAI_EMBEDDING"]]

Type of the embedding model.

class VertexAIEmbeddingConfig: …
component: Optional[VertexTextEmbedding]

Configuration for the VertexAI embedding model.

client_email: Optional[str]

The client email for the VertexAI credentials.

location: str

The default location to use when making API calls.

private_key: Optional[str]

The private key for the VertexAI credentials.

private_key_id: Optional[str]

The private key ID for the VertexAI credentials.

project: str

The default GCP project to use when making Vertex API calls.

token_uri: Optional[str]

The token URI for the VertexAI credentials.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the Vertex.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embed_mode: Optional[Literal["default", "classification", "clustering", 2 more]]

The embedding mode to use.

One of the following:
"default"
"classification"
"clustering"
"similarity"
"retrieval"
model_name: Optional[str]

The modelId of the VertexAI model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

type: Optional[Literal["VERTEXAI_EMBEDDING"]]

Type of the embedding model.

class BedrockEmbeddingConfig: …
component: Optional[BedrockEmbedding]

Configuration for the Bedrock embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the bedrock client.

aws_access_key_id: Optional[str]

AWS Access Key ID to use

aws_secret_access_key: Optional[str]

AWS Secret Access Key to use

aws_session_token: Optional[str]

AWS Session Token to use

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

The maximum number of API retries.

exclusiveMinimum0
model_name: Optional[str]

The modelId of the Bedrock model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

profile_name: Optional[str]

The name of aws profile to use. If not given, then the default profile is used.

region_name: Optional[str]

AWS region name to use. Uses region configured in AWS CLI if not passed

timeout: Optional[float]

The timeout for the Bedrock API request in seconds. It will be used for both connect and read timeouts.

type: Optional[Literal["BEDROCK_EMBEDDING"]]

Type of the embedding model.

name: str
project_id: str
config_hash: Optional[ConfigHash]

Hashes for the configuration of a pipeline.

embedding_config_hash: Optional[str]

Hash of the embedding config.

parsing_config_hash: Optional[str]

Hash of the llama parse parameters.

transform_config_hash: Optional[str]

Hash of the transform config.

created_at: Optional[datetime]

Creation datetime

formatdate-time
data_sink: Optional[DataSink]

Schema for a data sink.

id: str

Unique identifier

formatuuid
component: Component

Component that implements the data sink

One of the following:
Dict[str, object]
class CloudPineconeVectorStore: …

Cloud Pinecone Vector Store.

This class is used to store the configuration for a Pinecone vector store, so that it can be created and used in LlamaCloud.

Args: api_key (str): API key for authenticating with Pinecone index_name (str): name of the Pinecone index namespace (optional[str]): namespace to use in the Pinecone index insert_kwargs (optional[dict]): additional kwargs to pass during insertion

api_key: str

The API key for authenticating with Pinecone

formatpassword
index_name: str
class_name: Optional[str]
insert_kwargs: Optional[Dict[str, object]]
namespace: Optional[str]
supports_nested_metadata_filters: Optional[Literal[true]]
class CloudPostgresVectorStore: …
database: str
embed_dim: int
host: str
password: str
port: int
schema_name: str
table_name: str
user: str
class_name: Optional[str]
hnsw_settings: Optional[PgVectorHnswSettings]

HNSW settings for PGVector.

distance_method: Optional[Literal["l2", "ip", "cosine", 3 more]]

The distance method to use.

One of the following:
"l2"
"ip"
"cosine"
"l1"
"hamming"
"jaccard"
ef_construction: Optional[int]

The number of edges to use during the construction phase.

minimum1

The number of edges to use during the search phase.

minimum1
m: Optional[int]

The number of bi-directional links created for each new element.

minimum1
vector_type: Optional[Literal["vector", "half_vec", "bit", "sparse_vec"]]

The type of vector to use.

One of the following:
"vector"
"half_vec"
"bit"
"sparse_vec"
perform_setup: Optional[bool]
supports_nested_metadata_filters: Optional[bool]
class CloudQdrantVectorStore: …

Cloud Qdrant Vector Store.

This class is used to store the configuration for a Qdrant vector store, so that it can be created and used in LlamaCloud.

Args: collection_name (str): name of the Qdrant collection url (str): url of the Qdrant instance api_key (str): API key for authenticating with Qdrant max_retries (int): maximum number of retries in case of a failure. Defaults to 3 client_kwargs (dict): additional kwargs to pass to the Qdrant client

api_key: str
collection_name: str
url: str
class_name: Optional[str]
client_kwargs: Optional[Dict[str, object]]
max_retries: Optional[int]
supports_nested_metadata_filters: Optional[Literal[true]]
class CloudAzureAISearchVectorStore: …

Cloud Azure AI Search Vector Store.

search_service_api_key: str
search_service_endpoint: str
class_name: Optional[str]
client_id: Optional[str]
client_secret: Optional[str]
embedding_dimension: Optional[int]
filterable_metadata_field_keys: Optional[Dict[str, object]]
index_name: Optional[str]
search_service_api_version: Optional[str]
supports_nested_metadata_filters: Optional[Literal[true]]
tenant_id: Optional[str]

Cloud MongoDB Atlas Vector Store.

This class is used to store the configuration for a MongoDB Atlas vector store, so that it can be created and used in LlamaCloud.

Args: mongodb_uri (str): URI for connecting to MongoDB Atlas db_name (str): name of the MongoDB database collection_name (str): name of the MongoDB collection vector_index_name (str): name of the MongoDB Atlas vector index fulltext_index_name (str): name of the MongoDB Atlas full-text index

class CloudMilvusVectorStore: …

Cloud Milvus Vector Store.

uri: str
token: Optional[str]
class_name: Optional[str]
collection_name: Optional[str]
embedding_dimension: Optional[int]
supports_nested_metadata_filters: Optional[bool]
class CloudAstraDBVectorStore: …

Cloud AstraDB Vector Store.

This class is used to store the configuration for an AstraDB vector store, so that it can be created and used in LlamaCloud.

Args: token (str): The Astra DB Application Token to use. api_endpoint (str): The Astra DB JSON API endpoint for your database. collection_name (str): Collection name to use. If not existing, it will be created. embedding_dimension (int): Length of the embedding vectors in use. keyspace (optional[str]): The keyspace to use. If not provided, ‘default_keyspace’

token: str

The Astra DB Application Token to use

formatpassword
api_endpoint: str

The Astra DB JSON API endpoint for your database

collection_name: str

Collection name to use. If not existing, it will be created

embedding_dimension: int

Length of the embedding vectors in use

class_name: Optional[str]
keyspace: Optional[str]

The keyspace to use. If not provided, ‘default_keyspace’

supports_nested_metadata_filters: Optional[Literal[true]]
name: str

The name of the data sink.

project_id: str
sink_type: Literal["PINECONE", "POSTGRES", "QDRANT", 4 more]
One of the following:
"PINECONE"
"POSTGRES"
"QDRANT"
"AZUREAI_SEARCH"
"MONGODB_ATLAS"
"MILVUS"
"ASTRA_DB"
created_at: Optional[datetime]

Creation datetime

formatdate-time
updated_at: Optional[datetime]

Update datetime

formatdate-time
embedding_model_config: Optional[EmbeddingModelConfig]

Schema for an embedding model config.

id: str

Unique identifier

formatuuid
embedding_config: EmbeddingModelConfigEmbeddingConfig

The embedding configuration for the embedding model config.

One of the following:
class AzureOpenAIEmbeddingConfig: …
component: Optional[AzureOpenAIEmbedding]

Configuration for the Azure OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for Azure deployment.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for Azure OpenAI API.

azure_deployment: Optional[str]

The Azure deployment to use.

azure_endpoint: Optional[str]

The Azure endpoint to use.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["AZURE_EMBEDDING"]]

Type of the embedding model.

class CohereEmbeddingConfig: …
component: Optional[CohereEmbedding]

Configuration for the Cohere embedding model.

api_key: Optional[str]

The Cohere API key.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embedding_type: Optional[str]

Embedding type. If not provided float embedding_type is used when needed.

input_type: Optional[str]

Model Input type. If not provided, search_document and search_query are used when needed.

model_name: Optional[str]

The modelId of the Cohere model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

truncate: Optional[str]

Truncation type - START/ END/ NONE

type: Optional[Literal["COHERE_EMBEDDING"]]

Type of the embedding model.

class GeminiEmbeddingConfig: …
component: Optional[GeminiEmbedding]

Configuration for the Gemini embedding model.

api_base: Optional[str]

API base to access the model. Defaults to None.

api_key: Optional[str]

API key to access the model. Defaults to None.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
model_name: Optional[str]

The modelId of the Gemini model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

output_dimensionality: Optional[int]

Optional reduced dimension for output embeddings. Supported by models/text-embedding-004 and newer (e.g. gemini-embedding-001). Not supported by models/embedding-001.

task_type: Optional[str]

The task for embedding model.

title: Optional[str]

Title is only applicable for retrieval_document tasks, and is used to represent a document title. For other tasks, title is invalid.

transport: Optional[str]

Transport to access the model. Defaults to None.

type: Optional[Literal["GEMINI_EMBEDDING"]]

Type of the embedding model.

class HuggingFaceInferenceAPIEmbeddingConfig: …
component: Optional[HuggingFaceInferenceAPIEmbedding]

Configuration for the HuggingFace Inference API embedding model.

token: Optional[Union[str, bool, null]]

Hugging Face token. Will default to the locally saved token. Pass token=False if you don’t want to send your token to the server.

One of the following:
str
bool
class_name: Optional[str]
cookies: Optional[Dict[str, str]]

Additional cookies to send to the server.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
headers: Optional[Dict[str, str]]

Additional headers to send to the server. By default only the authorization and user-agent headers are sent. Values in this dictionary will override the default values.

model_name: Optional[str]

Hugging Face model name. If None, the task will be used.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

pooling: Optional[Literal["cls", "mean", "last"]]

Enum of possible pooling choices with pooling behaviors.

One of the following:
"cls"
"mean"
"last"
query_instruction: Optional[str]

Instruction to prepend during query embedding.

task: Optional[str]

Optional task to pick Hugging Face’s recommended model, used when model_name is left as default of None.

text_instruction: Optional[str]

Instruction to prepend during text embedding.

timeout: Optional[float]

The maximum number of seconds to wait for a response from the server. Loading a new model in Inference API can take up to several minutes. Defaults to None, meaning it will loop until the server is available.

type: Optional[Literal["HUGGINGFACE_API_EMBEDDING"]]

Type of the embedding model.

class OpenAIEmbeddingConfig: …
component: Optional[OpenAIEmbedding]

Configuration for the OpenAI embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the OpenAI API.

api_base: Optional[str]

The base URL for OpenAI API.

api_key: Optional[str]

The OpenAI API key.

api_version: Optional[str]

The version for OpenAI API.

class_name: Optional[str]
default_headers: Optional[Dict[str, str]]

The default headers for API requests.

dimensions: Optional[int]

The number of dimensions on the output embedding vectors. Works only with v3 embedding models.

embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

Maximum number of retries.

minimum0
model_name: Optional[str]

The name of the OpenAI embedding model.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

reuse_client: Optional[bool]

Reuse the OpenAI client between requests. When doing anything with large volumes of async API calls, setting this to false can improve stability.

timeout: Optional[float]

Timeout for each request.

minimum0
type: Optional[Literal["OPENAI_EMBEDDING"]]

Type of the embedding model.

class VertexAIEmbeddingConfig: …
component: Optional[VertexTextEmbedding]

Configuration for the VertexAI embedding model.

client_email: Optional[str]

The client email for the VertexAI credentials.

location: str

The default location to use when making API calls.

private_key: Optional[str]

The private key for the VertexAI credentials.

private_key_id: Optional[str]

The private key ID for the VertexAI credentials.

project: str

The default GCP project to use when making Vertex API calls.

token_uri: Optional[str]

The token URI for the VertexAI credentials.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the Vertex.

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
embed_mode: Optional[Literal["default", "classification", "clustering", 2 more]]

The embedding mode to use.

One of the following:
"default"
"classification"
"clustering"
"similarity"
"retrieval"
model_name: Optional[str]

The modelId of the VertexAI model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

type: Optional[Literal["VERTEXAI_EMBEDDING"]]

Type of the embedding model.

class BedrockEmbeddingConfig: …
component: Optional[BedrockEmbedding]

Configuration for the Bedrock embedding model.

additional_kwargs: Optional[Dict[str, object]]

Additional kwargs for the bedrock client.

aws_access_key_id: Optional[str]

AWS Access Key ID to use

aws_secret_access_key: Optional[str]

AWS Secret Access Key to use

aws_session_token: Optional[str]

AWS Session Token to use

class_name: Optional[str]
embed_batch_size: Optional[int]

The batch size for embedding calls.

maximum2048
exclusiveMinimum0
max_retries: Optional[int]

The maximum number of API retries.

exclusiveMinimum0
model_name: Optional[str]

The modelId of the Bedrock model to use.

num_workers: Optional[int]

The number of workers to use for async embedding calls.

profile_name: Optional[str]

The name of aws profile to use. If not given, then the default profile is used.

region_name: Optional[str]

AWS region name to use. Uses region configured in AWS CLI if not passed

timeout: Optional[float]

The timeout for the Bedrock API request in seconds. It will be used for both connect and read timeouts.

type: Optional[Literal["BEDROCK_EMBEDDING"]]

Type of the embedding model.

name: str

The name of the embedding model config.

project_id: str
created_at: Optional[datetime]

Creation datetime

formatdate-time
updated_at: Optional[datetime]

Update datetime

formatdate-time
embedding_model_config_id: Optional[str]

The ID of the EmbeddingModelConfig this pipeline is using.

formatuuid
llama_parse_parameters: Optional[LlamaParseParameters]

Settings that can be configured for how to use LlamaParse to parse files within a LlamaCloud pipeline.

adaptive_long_table: Optional[bool]
aggressive_table_extraction: Optional[bool]
auto_mode: Optional[bool]
auto_mode_configuration_json: Optional[str]
auto_mode_trigger_on_image_in_page: Optional[bool]
auto_mode_trigger_on_regexp_in_page: Optional[str]
auto_mode_trigger_on_table_in_page: Optional[bool]
auto_mode_trigger_on_text_in_page: Optional[str]
azure_openai_api_version: Optional[str]
azure_openai_deployment_name: Optional[str]
azure_openai_endpoint: Optional[str]
azure_openai_key: Optional[str]
bbox_bottom: Optional[float]
bbox_left: Optional[float]
bbox_right: Optional[float]
bbox_top: Optional[float]
bounding_box: Optional[str]
compact_markdown_table: Optional[bool]
complemental_formatting_instruction: Optional[str]
content_guideline_instruction: Optional[str]
continuous_mode: Optional[bool]
disable_image_extraction: Optional[bool]
disable_ocr: Optional[bool]
disable_reconstruction: Optional[bool]
do_not_cache: Optional[bool]
do_not_unroll_columns: Optional[bool]
enable_cost_optimizer: Optional[bool]
extract_charts: Optional[bool]
extract_layout: Optional[bool]
extract_printed_page_number: Optional[bool]
fast_mode: Optional[bool]
formatting_instruction: Optional[str]
gpt4o_api_key: Optional[str]
gpt4o_mode: Optional[bool]
guess_xlsx_sheet_name: Optional[bool]
hide_footers: Optional[bool]
hide_headers: Optional[bool]
high_res_ocr: Optional[bool]
html_make_all_elements_visible: Optional[bool]
html_remove_fixed_elements: Optional[bool]
html_remove_navigation_elements: Optional[bool]
http_proxy: Optional[str]
ignore_document_elements_for_layout_detection: Optional[bool]
images_to_save: Optional[List[Literal["screenshot", "embedded", "layout"]]]
One of the following:
"screenshot"
"embedded"
"layout"
inline_images_in_markdown: Optional[bool]
input_s3_path: Optional[str]
input_s3_region: Optional[str]
input_url: Optional[str]
internal_is_screenshot_job: Optional[bool]
invalidate_cache: Optional[bool]
is_formatting_instruction: Optional[bool]
job_timeout_extra_time_per_page_in_seconds: Optional[float]
job_timeout_in_seconds: Optional[float]
keep_page_separator_when_merging_tables: Optional[bool]
languages: Optional[List[ParsingLanguages]]
One of the following:
"af"
"az"
"bs"
"cs"
"cy"
"da"
"de"
"en"
"es"
"et"
"fr"
"ga"
"hr"
"hu"
"id"
"is"
"it"
"ku"
"la"
"lt"
"lv"
"mi"
"ms"
"mt"
"nl"
"no"
"oc"
"pi"
"pl"
"pt"
"ro"
"rs_latin"
"sk"
"sl"
"sq"
"sv"
"sw"
"tl"
"tr"
"uz"
"vi"
"ar"
"fa"
"ug"
"ur"
"bn"
"as"
"mni"
"ru"
"rs_cyrillic"
"be"
"bg"
"uk"
"mn"
"abq"
"ady"
"kbd"
"ava"
"dar"
"inh"
"che"
"lbe"
"lez"
"tab"
"tjk"
"hi"
"mr"
"ne"
"bh"
"mai"
"ang"
"bho"
"mah"
"sck"
"new"
"gom"
"sa"
"bgc"
"th"
"ch_sim"
"ch_tra"
"ja"
"ko"
"ta"
"te"
"kn"
layout_aware: Optional[bool]
line_level_bounding_box: Optional[bool]
markdown_table_multiline_header_separator: Optional[str]
max_pages: Optional[int]
max_pages_enforced: Optional[int]
merge_tables_across_pages_in_markdown: Optional[bool]
model: Optional[str]
outlined_table_extraction: Optional[bool]
output_pdf_of_document: Optional[bool]
output_s3_path_prefix: Optional[str]
output_s3_region: Optional[str]
output_tables_as_html: Optional[bool]
page_error_tolerance: Optional[float]
page_footer_prefix: Optional[str]
page_footer_suffix: Optional[str]
page_header_prefix: Optional[str]
page_header_suffix: Optional[str]
page_prefix: Optional[str]
page_separator: Optional[str]
page_suffix: Optional[str]
parse_mode: Optional[ParsingMode]

Enum for representing the mode of parsing to be used.

One of the following:
"parse_page_without_llm"
"parse_page_with_llm"
"parse_page_with_lvm"
"parse_page_with_agent"
"parse_page_with_layout_agent"
"parse_document_with_llm"
"parse_document_with_lvm"
"parse_document_with_agent"
parsing_instruction: Optional[str]
precise_bounding_box: Optional[bool]
premium_mode: Optional[bool]
presentation_out_of_bounds_content: Optional[bool]
presentation_skip_embedded_data: Optional[bool]
preserve_layout_alignment_across_pages: Optional[bool]
preserve_very_small_text: Optional[bool]
preset: Optional[str]
priority: Optional[Literal["low", "medium", "high", "critical"]]

The priority for the request. This field may be ignored or overwritten depending on the organization tier.

One of the following:
"low"
"medium"
"high"
"critical"
project_id: Optional[str]
remove_hidden_text: Optional[bool]
replace_failed_page_mode: Optional[FailPageMode]

Enum for representing the different available page error handling modes.

One of the following:
"raw_text"
"blank_page"
"error_message"
replace_failed_page_with_error_message_prefix: Optional[str]
replace_failed_page_with_error_message_suffix: Optional[str]
save_images: Optional[bool]
skip_diagonal_text: Optional[bool]
specialized_chart_parsing_agentic: Optional[bool]
specialized_chart_parsing_efficient: Optional[bool]
specialized_chart_parsing_plus: Optional[bool]
specialized_image_parsing: Optional[bool]
spreadsheet_extract_sub_tables: Optional[bool]
spreadsheet_force_formula_computation: Optional[bool]
spreadsheet_include_hidden_sheets: Optional[bool]
strict_mode_buggy_font: Optional[bool]
strict_mode_image_extraction: Optional[bool]
strict_mode_image_ocr: Optional[bool]
strict_mode_reconstruction: Optional[bool]
structured_output: Optional[bool]
structured_output_json_schema: Optional[str]
structured_output_json_schema_name: Optional[str]
system_prompt: Optional[str]
system_prompt_append: Optional[str]
take_screenshot: Optional[bool]
target_pages: Optional[str]
tier: Optional[str]
use_vendor_multimodal_model: Optional[bool]
user_prompt: Optional[str]
vendor_multimodal_api_key: Optional[str]
vendor_multimodal_model_name: Optional[str]
version: Optional[str]
webhook_configurations: Optional[List[WebhookConfiguration]]

Outbound webhook endpoints to notify on job status changes

webhook_events: Optional[List[Literal["extract.pending", "extract.success", "extract.error", 14 more]]]

Events to subscribe to (e.g. ‘parse.success’, ‘extract.error’). If null, all events are delivered.

One of the following:
"extract.pending"
"extract.success"
"extract.error"
"extract.partial_success"
"extract.cancelled"
"parse.pending"
"parse.running"
"parse.success"
"parse.error"
"parse.partial_success"
"parse.cancelled"
"classify.pending"
"classify.success"
"classify.error"
"classify.partial_success"
"classify.cancelled"
"unmapped_event"
webhook_headers: Optional[Dict[str, str]]

Custom HTTP headers sent with each webhook request (e.g. auth tokens)

webhook_output_format: Optional[str]

Response format sent to the webhook: ‘string’ (default) or ‘json’

webhook_url: Optional[str]

URL to receive webhook POST notifications

webhook_url: Optional[str]
managed_pipeline_id: Optional[str]

The ID of the ManagedPipeline this playground pipeline is linked to.

formatuuid
metadata_config: Optional[PipelineMetadataConfig]

Metadata configuration for the pipeline.

excluded_embed_metadata_keys: Optional[List[str]]

List of metadata keys to exclude from embeddings

excluded_llm_metadata_keys: Optional[List[str]]

List of metadata keys to exclude from LLM during retrieval

pipeline_type: Optional[PipelineType]

Type of pipeline. Either PLAYGROUND or MANAGED.

One of the following:
"PLAYGROUND"
"MANAGED"
preset_retrieval_parameters: Optional[PresetRetrievalParams]

Preset retrieval parameters for the pipeline.

alpha: Optional[float]

Alpha value for hybrid retrieval to determine the weights between dense and sparse retrieval. 0 is sparse retrieval and 1 is dense retrieval.

maximum1
minimum0
class_name: Optional[str]
dense_similarity_cutoff: Optional[float]

Minimum similarity score wrt query for retrieval

maximum1
minimum0
dense_similarity_top_k: Optional[int]

Number of nodes for dense retrieval.

maximum100
minimum1
enable_reranking: Optional[bool]

Enable reranking for retrieval

files_top_k: Optional[int]

Number of files to retrieve (only for retrieval mode files_via_metadata and files_via_content).

maximum5
minimum1
rerank_top_n: Optional[int]

Number of reranked nodes for returning.

maximum100
minimum1
retrieval_mode: Optional[RetrievalMode]

The retrieval mode for the query.

One of the following:
"chunks"
"files_via_metadata"
"files_via_content"
"auto_routed"
Deprecatedretrieve_image_nodes: Optional[bool]

Whether to retrieve image nodes.

retrieve_page_figure_nodes: Optional[bool]

Whether to retrieve page figure nodes.

retrieve_page_screenshot_nodes: Optional[bool]

Whether to retrieve page screenshot nodes.

search_filters: Optional[MetadataFilters]

Metadata filters for vector stores.

filters: List[Filter]
One of the following:
class FilterMetadataFilter: …

Comprehensive metadata filter for vector stores to support more operators.

Value uses Strict types, as int, float and str are compatible types and were all converted to string before.

See: https://docs.pydantic.dev/latest/usage/types/#strict-types

key: str
value: Union[float, str, List[str], 3 more]
One of the following:
float
str
List[str]
List[float]
List[int]
operator: Optional[Literal["==", ">", "<", 11 more]]

Vector store filter operator.

One of the following:
"=="
">"
"<"
"!="
">="
"<="
"in"
"nin"
"any"
"all"
"text_match"
"text_match_insensitive"
"contains"
"is_empty"
condition: Optional[Literal["and", "or", "not"]]

Vector store filter conditions to combine different filters.

One of the following:
"and"
"or"
"not"
search_filters_inference_schema: Optional[Dict[str, Union[Dict[str, object], List[object], str, 3 more]]]

JSON Schema that will be used to infer search_filters. Omit or leave as null to skip inference.

One of the following:
Dict[str, object]
List[object]
str
float
bool
sparse_similarity_top_k: Optional[int]

Number of nodes for sparse retrieval.

maximum100