Python Client API Reference

onecontext

OneContext

OneContext(api_key=None, base_url=None)

Initialize the OneContext client with an API key and base URL.

PARAMETER	DESCRIPTION
`api_key`	The API key for authenticating requests to the OneContext API. If not provided, the environment variable 'ONECONTEXT_API_KEY' will be used. Defaults to None. TYPE: `Optional[str]` DEFAULT: `None`
`base_url`	The base URL for the OneContext API. Defaults to "https://api.onecontext.ai/v1/". Can be overridden with the environment variable 'ONECONTEXT_API_KEY' TYPE: `str` DEFAULT: `None`

RAISES	DESCRIPTION
`ConfigurationError`	If no API key is provided and 'ONECONTEXT_API_KEY' environment variable is not set, a ConfigurationError is raised.

create_knowledgebase

create_knowledgebase(name)

Create a new knowledge base with the given name.

PARAMETER	DESCRIPTION
`name`	The name for the new knowledge base. TYPE: `str`

RETURNS	DESCRIPTION
`KnowledgeBase`	An instance of the KnowledgeBase class initialized with the given name.

delete_knowledgebase

delete_knowledgebase(name)

Delete a knowledge base by its name.

PARAMETER	DESCRIPTION
`name`	The name of the knowledge base to delete. TYPE: `str`

RETURNS	DESCRIPTION
`None`

list_knowledgebases

list_knowledgebases()

List the available Knowledge Bases.

This method fetches a list of knowledge base dictionaries from the API and instantiates a list of KnowledgeBase objects with the provided information.

RETURNS	DESCRIPTION
`List[KnowledgeBase]`	A list of KnowledgeBase objects representing the available knowledge bases.

KnowledgeBase

KnowledgeBase(name)

Factory method to create a KnowledgeBase object with the given name.

PARAMETER	DESCRIPTION
`name`	The name of the knowledge base to create. TYPE: `str`

RETURNS	DESCRIPTION
`KnowledgeBase`	An instance of KnowledgeBase associated with the provided name and the current client context.

deploy_pipeline

deploy_pipeline(name, pipeline_yaml_path=None, pipeline_yaml=None)

Deploys a pipeline based on the provided YAML configuration.

PARAMETER	DESCRIPTION
`name`	The name of the pipeline to be deployed. TYPE: `str`
`pipeline_yaml_path`	The file path to the pipeline YAML configuration. It can be a string or a Path object. Provide pipeline_yaml_path or pipeline_yaml, not both. TYPE: `Optional[Union[Path, str]]` DEFAULT: `None`
`pipeline_yaml`	The pipeline YAML configuration. Provide pipeline_yaml_path or pipeline_yaml, not both. TYPE: `Optional[str]` DEFAULT: `None`

RETURNS	DESCRIPTION
`Pipeline`	An instance of the deployed Pipeline.

RAISES	DESCRIPTION
`ValueError`	If the provided file path does not have a .yaml or .yml extension.

delete_pipeline

delete_pipeline(name)

Deletes a pipeline by name.

PARAMETER	DESCRIPTION
`name`	The name of the pipeline to delete. TYPE: `str`

RETURNS	DESCRIPTION
`None`

list_pipelines

list_pipelines()

Retrieve a list of Pipeline objects from the API.

RETURNS	DESCRIPTION
`List[Pipeline]`	A list of Pipeline objects.

list_runs

list_runs(*, status=None, run_id=None, skip=0, limit=20, sort='date_created', date_created_gte=None, date_created_lte=None)

Retrieve a list of runs with optional filtering and sorting.

PARAMETER	DESCRIPTION
`status`	The status of the runs to filter by. TYPE: `str` DEFAULT: `None`
`run_id`	The unique identifier for a run to filter by. TYPE: `str` DEFAULT: `None`
`skip`	The number of items to skip (for pagination), default is 0. TYPE: `int` DEFAULT: `0`
`limit`	The maximum number of items to return, default is 20. TYPE: `int` DEFAULT: `20`
`sort`	The field to sort the results by, default is 'date_created'. You can inverte the sort like so '-date_created' TYPE: `str` DEFAULT: `'date_created'`
`date_created_gte`	The minimum creation date to filter by (greater than or equal to). ISO 8601 Date Format Example: "2023-01-20T13:01:02Z" TYPE: `str` DEFAULT: `None`
`date_created_lte`	The maximum creation date to filter by (less than or equal to). ISO 8601 Date Format Example: "2023-01-20T13:01:02Z" TYPE: `str` DEFAULT: `None`

RETURNS	DESCRIPTION
`List[Dict[str, Any]]`	A list of dictionaries containing the run information.

Pipeline

Pipeline(name)

Create a new Pipeline object with a specified name.

Use this to access existing pipelines.

PARAMETER	DESCRIPTION
`name`	The name of the pipeline to create. TYPE: `str`

RETURNS	DESCRIPTION
`Pipeline`	An instance of the Pipeline class initialized with the given name.

create_vector_index

create_vector_index(name, model)

Creates a new vector index with the specified name and model.

PARAMETER	DESCRIPTION
`name`	The name of the index to be created. TYPE: `str`
`model`	The name of the model to use for the index. TYPE: `str`

RETURNS	DESCRIPTION
`VectorIndex`	An instance of `VectorIndex` with the response data.

Examples:

>>> oc = OneContextContext(...)
>>> index = oc.create_vector_index('my_index', 'BAAI/bge-base-en-v1.5')

delete_vector_index

delete_vector_index(name)

Delete a vecctor index base by its name.

PARAMETER	DESCRIPTION
`name`	The name of the vector index to delete. TYPE: `str`

RETURNS	DESCRIPTION
`None`

list_vector_indexes

list_vector_indexes()

Retrieve a list of VectorIndex objects from teh API.

RETURNS	DESCRIPTION
`list of VectorIndex`	A list of VectorIndex objects, each initialized with the index data and the reference to the current client and urls objects.

VectorIndex

VectorIndex(name, model_name=None)

Create a new instance of VectorIndex with the specified name and model name.

PARAMETER	DESCRIPTION
`name`	The name of the VectorIndex. TYPE: `str`
`model_name`	The name of the model associated with the VectorIndex, by default None. TYPE: `Optional[str]` DEFAULT: `None`

RETURNS	DESCRIPTION
`VectorIndex`	An instance of VectorIndex with the given configuration.

set_openai_key

set_openai_key(openai_api_key)

Submit your OpenAI API key to OneContext

OneContext encrypts your key with a symmetric Google KMS key and only the ciphertext is stored on our servers.

You can learn more about symmetric Google KMS encryption here: https://cloud.google.com/kms/docs/encrypt-decrypt

PARAMETER	DESCRIPTION
`openai_api_key`	The OpenAI API key to be used by the client. TYPE: `str`

Chunk `dataclass`

Chunk(id, content, metadata_json=None, file_name=None, date_created=None)

A chunk of a file, stored in a vector index

ATTRIBUTE	DESCRIPTION
`id`	A unique identifier for the chunk. TYPE: `str`
`content`	The actual content of the chunk. TYPE: `str`
`metadata_json`	A dictionary containing metadata about the chunk. TYPE: `(Optional[dict], optional)`
`file_name`	The name of the file associated with the chunk. TYPE: `(str \| None, optional)`

chunk

Chunk `dataclass`

Chunk(id, content, metadata_json=None, file_name=None, date_created=None)

A chunk of a file, stored in a vector index

ATTRIBUTE	DESCRIPTION
`id`	A unique identifier for the chunk. TYPE: `str`
`content`	The actual content of the chunk. TYPE: `str`
`metadata_json`	A dictionary containing metadata about the chunk. TYPE: `(Optional[dict], optional)`
`file_name`	The name of the file associated with the chunk. TYPE: `(str \| None, optional)`

index

VectorIndex `dataclass`

VectorIndex(name, model_name, _client, _urls)

A dataclass that represents a vector index in the OneContext API.

PARAMETER	DESCRIPTION
`name`	The name of the vector index. TYPE: `str`
`model_name`	The name of the model associated with this vector index. TYPE: `str`
`_client`	An instance of the ApiClient. TYPE: `ApiClient`
`_urls`	An instance of URLS containing the API endpoints. TYPE: `URLS`

list_chunks

list_chunks(file_names=None, file_ids=None, chunk_ids=None, skip=0, limit=200, sort='date_created', metadata_filters=None, date_created_gte=None, date_created_lte=None)

Lists chunks in the knowledge base with various filtering, sorting, and pagination options.

PARAMETER	DESCRIPTION
`file_names`	A list of names of the files to filter results (default is None). TYPE: `list[str]` DEFAULT: `None`
`file_ids`	A list of IDs of the files to filter results (default is None). TYPE: `list[str]` DEFAULT: `None`
`chunk_ids`	A list of IDs of the chunks to filter results (default is None). TYPE: `list[str]` DEFAULT: `None`
`skip`	The number of chunks to skip (default is 0). TYPE: `int` DEFAULT: `0`
`limit`	The maximum number of chunks to return (default is 200). TYPE: `int` DEFAULT: `200`
`sort`	The field to sort by (default is "date_created"). Reverse with "-date_created" TYPE: `str` DEFAULT: `'date_created'`
`metadata_filters`	A dictionary of metadata fields to filter results (default is None). TYPE: `dict` DEFAULT: `None`
`date_created_gte`	The minimum creation date of chunks to list (default is None). ISO 8601 Date Format Example: "2023-01-20T13:01:02Z" TYPE: `datetime` DEFAULT: `None`
`date_created_lte`	The maximum creation date of chunks to list (default is None). ISO 8601 Date Format Example: "2023-01-20T13:01:02Z" TYPE: `datetime` DEFAULT: `None`

RETURNS	DESCRIPTION
`List[Chunk]`	A list of Chunk objects that are the result of running the query

list_files

list_files(skip=0, limit=500, sort='date_created', metadata_filters=None, date_created_gte=None, date_created_lte=None)

Lists files associated to the chunks in the vecotr index.

PARAMETER	DESCRIPTION
`skip`	The number of files to skip (default is 0). TYPE: `int` DEFAULT: `0`
`limit`	The maximum number of files to return (default is 20). TYPE: `int` DEFAULT: `500`
`sort`	The field to sort by (default is "date_created"). Reverse with "-date_created" TYPE: `str` DEFAULT: `'date_created'`
`metadata_filters`	A dictionary of metadata fields to filter results (default is None). TYPE: `dict` DEFAULT: `None`
`date_created_gte`	The minimum creation date of files to list (default is None). ISO 8601 Date Format Example: "2023-01-20T13:01:02Z" TYPE: `datetime` DEFAULT: `None`
`date_created_lte`	The maximum creation date of files to list (default is None). ISO 8601 Date Format Example: "2023-01-20T13:01:02Z" TYPE: `datetime` DEFAULT: `None`

RETURNS	DESCRIPTION
`List[Dict[str, Any]]`	A list of dictionaries, each representing a file with its metadata.

knowledgebase

KnowledgeBase `dataclass`

KnowledgeBase(name, _client, _urls, id=None)

A class representing a knowledge base client to interact with an API for file management.

ATTRIBUTE	DESCRIPTION
`name`	The name of the knowledge base. TYPE: `str`
`_client`	The API client to make requests to the knowledge base. TYPE: `ApiClient`
`_urls`	Object containing URL templates for API requests. TYPE: `URLS`
`id`	The identifier for the knowledge base, by default None. TYPE: `(Optional[str], optional)`

list_files

list_files(skip=0, limit=500, sort='date_created', metadata_filters=None, date_created_gte=None, date_created_lte=None)

Lists files in the knowledge base with various filtering, sorting, and pagination options.

PARAMETER	DESCRIPTION
`skip`	The number of files to skip (default is 0). TYPE: `int` DEFAULT: `0`
`limit`	The maximum number of files to return (default is 20). TYPE: `int` DEFAULT: `500`
`sort`	The field to sort by (default is "date_created"). Reverse with "-date_created" TYPE: `str` DEFAULT: `'date_created'`
`metadata_filters`	A dictionary of metadata fields to filter results (default is None). TYPE: `dict` DEFAULT: `None`
`date_created_gte`	The minimum creation date of files to list (default is None). ISO 8601 Date Format Example: "2023-01-20T13:01:02Z" TYPE: `datetime` DEFAULT: `None`
`date_created_lte`	The maximum creation date of files to list (default is None). ISO 8601 Date Format Example: "2023-01-20T13:01:02Z" TYPE: `datetime` DEFAULT: `None`

RETURNS	DESCRIPTION
`List[Dict[str, Any]]`	A list of dictionaries, each representing a file with its metadata.

delete_files

delete_files(file_names)

Deletes a list of files from the knowledge base.

PARAMETER	DESCRIPTION
`file_names`	A list of filenames as strings that are to be deleted. TYPE: `list[str]`

RETURNS	DESCRIPTION
`None`

delete_duplicate_files

delete_duplicate_files()

Deletes duplicate files by file_name in the knowledge_base

The most recent file for each duplicate file_name is retained

RETURNS	DESCRIPTION
`None`

upload_file

upload_file(file_path, metadata=None)

Uploads a file to the knowledge base.

This method uploads a file specified by file_path to the knowledge base, optionally associating it with metadata.

PARAMETER	DESCRIPTION
`file_path`	The path to the file to be uploaded. Can be a string or a Path object. TYPE: `Union[str, Path]`
`metadata`	A dictionary containing metadata for the file. The keys "file_name", "knowledge_base", "user_id", "file_path", and "file_id" are reserved and cannot be used. If provided, it should not contain any of these keys. The default is None, which means no metadata will be associated with the file. TYPE: `Optional[dict]` DEFAULT: `None`

RETURNS	DESCRIPTION
`list[str]`	A list of pipeline run IDs generated after the file is uploaded.

RAISES	DESCRIPTION
`ValueError`	If any reserved keys are present in the metadata. If the file type is not supported (not in SUPPORTED_FILE_TYPES).

upload_text

upload_text(text, file_name, metadata=None)

Uploads a text string as a file to a knowledge base with optional metadata.

PARAMETER	DESCRIPTION
`text`	The text content to be uploaded. TYPE: `str`
`file_name`	The name of the file to be created. If the provided file_name does not end with '.txt', it is automatically modified to end with this extension. TYPE: `str`
`metadata`	A dictionary containing metadata for the upload. There are reserved keys that cannot be used: 'file_name', 'knowledge_base', 'user_id', 'file_path', and 'file_id'. If any of these keys are present, a ValueError is raised. TYPE: `dict` DEFAULT: `None`

RAISES	DESCRIPTION
`ValueError`	If any reserved key is present in the metadata dictionary.

upload_from_directory

upload_from_directory(directory, metadata=None)

Uploads files from a given directory to a knowledgebase.

This method uploads all files within a specified directory that match the supported file types.

PARAMETER	DESCRIPTION
`directory`	The path to the directory containing the files to be uploaded. Can be a string or a Path object. TYPE: `Union[str, Path]`
`metadata`	Metadata associated with the files to be uploaded. This can be a single dictionary applied to all files, or a list of dictionaries with one dictionary for each file. If a list is provided, its length must match the number of files. If omitted or `None`, no metadata will be applied. TYPE: `Optional[Union[dict, List[dict]]]` DEFAULT: `None`

RAISES	DESCRIPTION
`ValueError`	If the provided directory is not an actual directory or if the length of the metadata list does not match the number of files to be uploaded.

main

OneContext

OneContext(api_key=None, base_url=None)

Initialize the OneContext client with an API key and base URL.

PARAMETER	DESCRIPTION
`api_key`	The API key for authenticating requests to the OneContext API. If not provided, the environment variable 'ONECONTEXT_API_KEY' will be used. Defaults to None. TYPE: `Optional[str]` DEFAULT: `None`
`base_url`	The base URL for the OneContext API. Defaults to "https://api.onecontext.ai/v1/". Can be overridden with the environment variable 'ONECONTEXT_API_KEY' TYPE: `str` DEFAULT: `None`

RAISES	DESCRIPTION
`ConfigurationError`	If no API key is provided and 'ONECONTEXT_API_KEY' environment variable is not set, a ConfigurationError is raised.

create_knowledgebase

create_knowledgebase(name)

Create a new knowledge base with the given name.

PARAMETER	DESCRIPTION
`name`	The name for the new knowledge base. TYPE: `str`

RETURNS	DESCRIPTION
`KnowledgeBase`	An instance of the KnowledgeBase class initialized with the given name.

delete_knowledgebase

delete_knowledgebase(name)

Delete a knowledge base by its name.

PARAMETER	DESCRIPTION
`name`	The name of the knowledge base to delete. TYPE: `str`

RETURNS	DESCRIPTION
`None`

list_knowledgebases

list_knowledgebases()

List the available Knowledge Bases.

This method fetches a list of knowledge base dictionaries from the API and instantiates a list of KnowledgeBase objects with the provided information.

RETURNS	DESCRIPTION
`List[KnowledgeBase]`	A list of KnowledgeBase objects representing the available knowledge bases.

KnowledgeBase

KnowledgeBase(name)

Factory method to create a KnowledgeBase object with the given name.

PARAMETER	DESCRIPTION
`name`	The name of the knowledge base to create. TYPE: `str`

RETURNS	DESCRIPTION
`KnowledgeBase`	An instance of KnowledgeBase associated with the provided name and the current client context.

deploy_pipeline

deploy_pipeline(name, pipeline_yaml_path=None, pipeline_yaml=None)

Deploys a pipeline based on the provided YAML configuration.

PARAMETER	DESCRIPTION
`name`	The name of the pipeline to be deployed. TYPE: `str`
`pipeline_yaml_path`	The file path to the pipeline YAML configuration. It can be a string or a Path object. Provide pipeline_yaml_path or pipeline_yaml, not both. TYPE: `Optional[Union[Path, str]]` DEFAULT: `None`
`pipeline_yaml`	The pipeline YAML configuration. Provide pipeline_yaml_path or pipeline_yaml, not both. TYPE: `Optional[str]` DEFAULT: `None`

RETURNS	DESCRIPTION
`Pipeline`	An instance of the deployed Pipeline.

RAISES	DESCRIPTION
`ValueError`	If the provided file path does not have a .yaml or .yml extension.

delete_pipeline

delete_pipeline(name)

Deletes a pipeline by name.

PARAMETER	DESCRIPTION
`name`	The name of the pipeline to delete. TYPE: `str`

RETURNS	DESCRIPTION
`None`

list_pipelines

list_pipelines()

Retrieve a list of Pipeline objects from the API.

RETURNS	DESCRIPTION
`List[Pipeline]`	A list of Pipeline objects.

list_runs

list_runs(*, status=None, run_id=None, skip=0, limit=20, sort='date_created', date_created_gte=None, date_created_lte=None)

Retrieve a list of runs with optional filtering and sorting.

PARAMETER	DESCRIPTION
`status`	The status of the runs to filter by. TYPE: `str` DEFAULT: `None`
`run_id`	The unique identifier for a run to filter by. TYPE: `str` DEFAULT: `None`
`skip`	The number of items to skip (for pagination), default is 0. TYPE: `int` DEFAULT: `0`
`limit`	The maximum number of items to return, default is 20. TYPE: `int` DEFAULT: `20`
`sort`	The field to sort the results by, default is 'date_created'. You can inverte the sort like so '-date_created' TYPE: `str` DEFAULT: `'date_created'`
`date_created_gte`	The minimum creation date to filter by (greater than or equal to). ISO 8601 Date Format Example: "2023-01-20T13:01:02Z" TYPE: `str` DEFAULT: `None`
`date_created_lte`	The maximum creation date to filter by (less than or equal to). ISO 8601 Date Format Example: "2023-01-20T13:01:02Z" TYPE: `str` DEFAULT: `None`

RETURNS	DESCRIPTION
`List[Dict[str, Any]]`	A list of dictionaries containing the run information.

Pipeline

Pipeline(name)

Create a new Pipeline object with a specified name.

Use this to access existing pipelines.

PARAMETER	DESCRIPTION
`name`	The name of the pipeline to create. TYPE: `str`

RETURNS	DESCRIPTION
`Pipeline`	An instance of the Pipeline class initialized with the given name.

create_vector_index

create_vector_index(name, model)

Creates a new vector index with the specified name and model.

PARAMETER	DESCRIPTION
`name`	The name of the index to be created. TYPE: `str`
`model`	The name of the model to use for the index. TYPE: `str`

RETURNS	DESCRIPTION
`VectorIndex`	An instance of `VectorIndex` with the response data.

Examples:

>>> oc = OneContextContext(...)
>>> index = oc.create_vector_index('my_index', 'BAAI/bge-base-en-v1.5')

delete_vector_index

delete_vector_index(name)

Delete a vecctor index base by its name.

PARAMETER	DESCRIPTION
`name`	The name of the vector index to delete. TYPE: `str`

RETURNS	DESCRIPTION
`None`

list_vector_indexes

list_vector_indexes()

Retrieve a list of VectorIndex objects from teh API.

RETURNS	DESCRIPTION
`list of VectorIndex`	A list of VectorIndex objects, each initialized with the index data and the reference to the current client and urls objects.

VectorIndex

VectorIndex(name, model_name=None)

Create a new instance of VectorIndex with the specified name and model name.

PARAMETER	DESCRIPTION
`name`	The name of the VectorIndex. TYPE: `str`
`model_name`	The name of the model associated with the VectorIndex, by default None. TYPE: `Optional[str]` DEFAULT: `None`

RETURNS	DESCRIPTION
`VectorIndex`	An instance of VectorIndex with the given configuration.

set_openai_key

set_openai_key(openai_api_key)

Submit your OpenAI API key to OneContext

OneContext encrypts your key with a symmetric Google KMS key and only the ciphertext is stored on our servers.

You can learn more about symmetric Google KMS encryption here: https://cloud.google.com/kms/docs/encrypt-decrypt

PARAMETER	DESCRIPTION
`openai_api_key`	The OpenAI API key to be used by the client. TYPE: `str`

pipeline

Chunk `dataclass`

Chunk(id, content, metadata_json=None, file_name=None, date_created=None)

A chunk of a file, stored in a vector index

ATTRIBUTE	DESCRIPTION
`id`	A unique identifier for the chunk. TYPE: `str`
`content`	The actual content of the chunk. TYPE: `str`
`metadata_json`	A dictionary containing metadata about the chunk. TYPE: `(Optional[dict], optional)`
`file_name`	The name of the file associated with the chunk. TYPE: `(str \| None, optional)`

Pipeline `dataclass`

Pipeline(_client, _urls, name, id=None, yaml_config=None, run_id=None)

A dataclass that represents a pipeline capable of interacting with an API client to perform various operations.

PARAMETER	DESCRIPTION
`name`	The name of the pipeline. TYPE: `str`
`id`	The unique identifier of the pipeline. TYPE: `Optional[str]` DEFAULT: `None`
`yaml_config`	The YAML configuration of the pipeline, if any. TYPE: `Optional[str]` DEFAULT: `None`
`run_results`	The results of the pipeline run, if any. TYPE: `Optional[str]` DEFAULT: `None`
`run_id`	The unique identifier of the run that is started when the pipeline is first deployed. TYPE: `Optional[str]` DEFAULT: `None`
`_client`	An instance of ApiClient to make API calls. This parameter is hidden in the representation of the object. TYPE: `ApiClient`
`_urls`	An instance of URLS that contains the endpoint URLs required for API calls. This parameter is hidden in the representation of the object. TYPE: `URLS`

run

run(override_args=None)

Runs the pipeline with optional override arguments.

PARAMETER	DESCRIPTION
`override_args`	A dictionary of step names and step_args to override the default pipeline arguments, by default None. TYPE: `dict[str, Any]` DEFAULT: `None`

RETURNS	DESCRIPTION
`List[Chunk]`	A list of Chunk objects that are the result of running the pipeline with the specified parameters.

Examples:

>>> query_pipeline = oc.Pipeline("basic_query")
>>> query = "What are consequences of inventing a computer?"
>>> retriever_top_k = 50
>>> top_k = 5
>>> override_args = {
...     "retriever": {
...         "top_k": retriever_top_k,
...         "query": query,
...     },
...     "reranker": {"top_k": top_k, "query": query},
... }
>>> chunks = query_pipeline.run(override_args)

Python Client API Reference

onecontext

OneContext

create_knowledgebase

delete_knowledgebase

list_knowledgebases

KnowledgeBase

deploy_pipeline

delete_pipeline

list_pipelines

list_runs

Pipeline

create_vector_index

delete_vector_index

list_vector_indexes

VectorIndex

set_openai_key

Chunk dataclass

chunk

Chunk dataclass

index

VectorIndex dataclass

list_chunks

list_files

knowledgebase

KnowledgeBase dataclass

list_files

delete_files

delete_duplicate_files

upload_file

upload_text

upload_from_directory

main

OneContext

create_knowledgebase

delete_knowledgebase

list_knowledgebases

KnowledgeBase

deploy_pipeline

delete_pipeline

list_pipelines

list_runs

Pipeline

create_vector_index

delete_vector_index

list_vector_indexes

VectorIndex

set_openai_key

pipeline

Chunk dataclass

Pipeline dataclass

run

Chunk `dataclass`

Chunk `dataclass`

VectorIndex `dataclass`

KnowledgeBase `dataclass`

Chunk `dataclass`

Pipeline `dataclass`