Vector Data client library for Python

FactSet's Vector Data API helps asset managers, hedge funds, investment banks, and financial technology firms accelerate the development of next-generation AI and machine learning solutions by simplifying and optimizing access to unstructured financial documents.

This API provides streamlined access to vector data through its defined endpoints. It supports retrieving detailed vector data based on user-defined parameters. Efficiently processing associated text data for enhanced performance.

This API is designed to enable developers to integrate vector data into their applications, ensuring flexibility and performance while leveraging the specified endpoint functionalities.

Access to the AI Ready version of FactSet collected documents with enrichments. Meant as an input to a client's RAG experience.

Note: Vector Data API performs semantic search across vectorized financial documents and returns results ranked by similarity score. Supports filtering by themes, sentiment, form types, date ranges, and entity identifiers.

API Build & Advantages:

AI-Ready Unstructured Data Processing Pipeline: Automate the ingestion, chunking, embedding, and indexing of high-value financial text-removing the need for manual pipeline development.

Out-of-the-box support for FactSet proprietary and third-party datasets (StreetAccount News, CallStreet Transcripts, EDGAR Filings, MT Newswire and more to come).

Benchmark-driven chunking strategies ensure optimal results for GenAI and RAG use.

Platform & Embedding Model Agnostic: Seamlessly integrate with any AI stack-works with cloud and on-prem environments and supports any embedding model you choose.

Native Integration with Model Context Protocol (MCP): Enable direct, standardized connectivity between your unstructured data workflows and MCP-compliant AI solutions, supporting context-aware generative AI, retrieval-augmented generation, and document-level search intelligence. Rich Metadata, Tagging, and Entity Resolution: Enhance AI models with entity and date resolution, topic tagging, sentiment labels, and links to FactSet's Unstructured Knowledge Graph, enabling richer, more accurate search and analysis.

This Python package is automatically generated by the OpenAPI Generator project:

• API version: 0.4.0
• SDK version: 0.2.0
• Build package: org.openapitools.codegen.languages.PythonClientCodegen

For more information, please visit https://developer.factset.com/contact

Requirements

• Python >= 3.7

Installation

Poetry

poetry add fds.sdk.utils fds.sdk.VectorData==0.2.0

pip

pip install fds.sdk.utils fds.sdk.VectorData==0.2.0

Usage

1. Generate authentication credentials.
2. Setup Python environment.

   1. Install and activate python 3.10+. If you're using pyenv:

      pyenv install 3.10.0
      pyenv shell 3.10.0
      

   2. (optional) Install poetry.

3. Install dependencies.
4. Run the following:

Example Code

from fds.sdk.utils.authentication import ConfidentialClient

import fds.sdk.VectorData
from fds.sdk.VectorData.api import intelligent_document_service_api
from fds.sdk.VectorData.models import *
from dateutil.parser import parse as dateutil_parser
from pprint import pprint

# See configuration.py for a list of all supported configuration parameters.

# Examples for each supported authentication method are below,
# choose one that satisfies your use case.

# (Preferred) OAuth 2.0: FactSetOAuth2
# See https://github.com/FactSet/enterprise-sdk#oauth-20
# for information on how to create the app-config.json file
#
# The confidential client instance should be reused in production environments.
# See https://github.com/FactSet/enterprise-sdk-utils-python#authentication
# for more information on using the ConfidentialClient class
configuration = fds.sdk.VectorData.Configuration(
    fds_oauth_client=ConfidentialClient('/path/to/app-config.json')
)

# Basic authentication: FactSetApiKey
# See https://github.com/FactSet/enterprise-sdk#api-key
# for information how to create an API key
# configuration = fds.sdk.VectorData.Configuration(
#     username='USERNAME-SERIAL',
#     password='API-KEY'
# )

# Enter a context with an instance of the API client
with fds.sdk.VectorData.ApiClient(configuration) as api_client:
    # Create an instance of the API class
    api_instance = intelligent_document_service_api.IntelligentDocumentServiceApi(api_client)
    ids_document_upload_request = IdsDocumentUploadRequest(
        data=IdsDocumentUploadRequestData(
            file_name="annual-report-2024.pdf",
            file_type="Pdf",
        ),
    ) # IdsDocumentUploadRequest | Document registration request.

    try:
        # Register a document and receive a presigned upload URL
        # example passing only required values which don't have defaults set
        api_response = api_instance.create_ids_document_upload(ids_document_upload_request)

        pprint(api_response)
    except fds.sdk.VectorData.ApiException as e:
        print("Exception when calling IntelligentDocumentServiceApi->create_ids_document_upload: %s\n" % e)

    # # Get response, http status code and response headers
    # try:
    #     # Register a document and receive a presigned upload URL
    #     api_response, http_status_code, response_headers = api_instance.create_ids_document_upload_with_http_info(ids_document_upload_request)


    #     pprint(api_response)
    #     pprint(http_status_code)
    #     pprint(response_headers)
    # except fds.sdk.VectorData.ApiException as e:
    #     print("Exception when calling IntelligentDocumentServiceApi->create_ids_document_upload: %s\n" % e)

    # # Get response asynchronous
    # try:
    #     # Register a document and receive a presigned upload URL
    #     async_result = api_instance.create_ids_document_upload_async(ids_document_upload_request)
    #     api_response = async_result.get()


    #     pprint(api_response)
    # except fds.sdk.VectorData.ApiException as e:
    #     print("Exception when calling IntelligentDocumentServiceApi->create_ids_document_upload: %s\n" % e)

    # # Get response, http status code and response headers asynchronous
    # try:
    #     # Register a document and receive a presigned upload URL
    #     async_result = api_instance.create_ids_document_upload_with_http_info_async(ids_document_upload_request)
    #     api_response, http_status_code, response_headers = async_result.get()


    #     pprint(api_response)
    #     pprint(http_status_code)
    #     pprint(response_headers)
    # except fds.sdk.VectorData.ApiException as e:
    #     print("Exception when calling IntelligentDocumentServiceApi->create_ids_document_upload: %s\n" % e)

Using Pandas

To convert an API response to a Pandas DataFrame, it is necessary to transform it first to a dictionary.

import pandas as pd

response_dict = api_response.to_dict()['data']

simple_json_response = pd.DataFrame(response_dict)
nested_json_response = pd.json_normalize(response_dict)

Debugging

The SDK uses the standard library logging module.

Setting debug to True on an instance of the Configuration class sets the log-level of related packages to DEBUG and enables additional logging in Pythons HTTP Client.

Note: This prints out sensitive information (e.g. the full request and response). Use with care.

import logging
import fds.sdk.VectorData

logging.basicConfig(level=logging.DEBUG)

configuration = fds.sdk.VectorData.Configuration(...)
configuration.debug = True

Configure a Proxy

You can pass proxy settings to the Configuration class:

• proxy: The URL of the proxy to use.
• proxy_headers: a dictionary to pass additional headers to the proxy (e.g. Proxy-Authorization).

import fds.sdk.VectorData

configuration = fds.sdk.VectorData.Configuration(
    # ...
    proxy="http://secret:password@localhost:5050",
    proxy_headers={
        "Custom-Proxy-Header": "Custom-Proxy-Header-Value"
    }
)

Custom SSL Certificate

TLS/SSL certificate verification can be configured with the following Configuration parameters:

• ssl_ca_cert: a path to the certificate to use for verification in PEM format.
• verify_ssl: setting this to False disables the verification of certificates. Disabling the verification is not recommended, but it might be useful during local development or testing.

import fds.sdk.VectorData

configuration = fds.sdk.VectorData.Configuration(
    # ...
    ssl_ca_cert='/path/to/ca.pem'
)

Request Retries

In case the request retry behaviour should be customized, it is possible to pass a urllib3.Retry object to the retry property of the Configuration.

from urllib3 import Retry
import fds.sdk.VectorData

configuration = fds.sdk.VectorData.Configuration(
    # ...
)

configuration.retries = Retry(total=3, status_forcelist=[500, 502, 503, 504])

Documentation for API Endpoints

All URIs are relative to https://api.factset.com/content/vector/v0

Class	Method	HTTP request	Description
IntelligentDocumentServiceApi	create_ids_document_upload	POST /documents/register	Register a document and receive a presigned upload URL
IntelligentDocumentServiceApi	get_ids_document_download	POST /documents/download	Get a presigned URL to download processed document output.
IntelligentDocumentServiceApi	get_ids_document_status	POST /documents/status	Retrieve ingestion status for one or more documents.
IntelligentDocumentServiceApi	list_all_ids_documents	GET /documents/list	List all uploaded documents
IntelligentDocumentServiceApi	search_ids_documents	POST /documents/search	Semantically search uploaded documents
VectorApi	get_count	GET /chunk-text	Returns chunked text for the given vectorId.
VectorApi	get_document_types	GET /meta/document-types	Returns the document types.
VectorApi	get_sources	GET /meta/sources	Returns the sources.
VectorApi	get_themes	GET /meta/themes	Returns the themes.
VectorApi	getschemas	GET /meta/schemas	Returns the schemas.
VectorApi	post_vector	POST /data	Return vector information based on the input parameters below
VectorSearchApi	get_form_types	GET /form-types	Returns the available form types
VectorSearchApi	get_vector	GET /vector	Returns vector data and metadata for specified vector IDs
VectorSearchApi	post_search	POST /search	Perform semantic search and return ranked results with similarity scores
VectorSearchApi	retrieve_sources	GET /sources	Returns the available data sources

Documentation For Models

• BaseResult
• ChunkTextResponse
• ChunkTextResponseMeta
• ChunkTextResponseMetaPagination
• ChunkTextResult
• DocumentTypes
• DocumentTypesResponse
• DocumentsDownload
• DocumentsStatus
• EDGARSearch
• EDGARSearchResult
• EDGARSearchResultAllOf
• EDGARVector
• EDGARVectorResult
• EDGARVectorResultAllOf
• ErrorObject
• ErrorObjectData
• ErrorObjectDataSource
• ErrorObjectResponse
• ErrorResponse
• ErrorResult
• FormTypes
• FormTypesResponse
• IdsDocumentDownloadData
• IdsDocumentDownloadRequest
• IdsDocumentDownloadRequestData
• IdsDocumentDownloadResponse
• IdsDocumentListItem
• IdsDocumentSearchRequest
• IdsDocumentSearchRequestData
• IdsDocumentSearchRequestMeta
• IdsDocumentSearchResponse
• IdsDocumentSearchResponseMeta
• IdsDocumentSearchResult
• IdsDocumentStatusData
• IdsDocumentStatusRequest
• IdsDocumentStatusRequestData
• IdsDocumentStatusResponse
• IdsDocumentUploadData
• IdsDocumentUploadRequest
• IdsDocumentUploadRequestData
• IdsDocumentUploadResponse
• IdsDocumentsListResponse
• Meta
• NewsSearch
• NewsSearchResult
• NewsSearchResultAllOf
• NewsVector
• NewsVectorResult
• NewsVectorResultAllOf
• OffsetPaginationRequest
• OffsetPaginationResponse
• Schemas
• SchemasResponse
• SearchRequest
• SearchRequestData
• SearchRequestMeta
• SearchRequestMetaPagination
• SearchResponse
• SearchResponseMeta
• SearchResponseMetaPagination
• SearchResult
• Source
• SourceData
• SourceResponse
• SourceResponseData
• Themes
• ThemesResponse
• TranscriptSearch
• TranscriptSearchResult
• TranscriptSearchResultAllOf
• TranscriptVector
• TranscriptVectorResult
• TranscriptVectorResultAllOf
• VectorDataRequest
• VectorDataRequestData
• VectorDataResponse
• VectorDataResponseMeta
• VectorDataResult
• VectorResponse
• VectorResult

Documentation For Authorization

FactSetApiKey

• Type: HTTP basic authentication

FactSetOAuth2

• Type: OAuth
• Flow: application
• Authorization URL:
• Scopes: N/A

Notes for Large OpenAPI documents

If the OpenAPI document is large, imports in fds.sdk.VectorData.apis and fds.sdk.VectorData.models may fail with a RecursionError indicating the maximum recursion limit has been exceeded. In that case, there are a couple of solutions:

Solution 1: Use specific imports for apis and models like:

• from fds.sdk.VectorData.api.default_api import DefaultApi
• from fds.sdk.VectorData.model.pet import Pet

Solution 2: Before importing the package, adjust the maximum recursion limit as shown below:

import sys
sys.setrecursionlimit(1500)
import fds.sdk.VectorData
from fds.sdk.VectorData.apis import *
from fds.sdk.VectorData.models import *

Contributing

Please refer to the contributing guide.

Copyright

Copyright 2026 FactSet Research Systems Inc

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.