AI Connect | Data Science Connect
AI ConnectExplore the 2027 portfolio

Data Infrastructure

Unstructured

Turns documents and unstructured files into AI-ready data

Visit website ↗See if it’s a matchRequest an introduction

About Unstructured

Unstructured offers an open-source library and a commercial platform that ingest, partition and transform documents such as PDFs, slides, emails and images into structured formats for LLM and RAG applications. It supports connectors for common sources and destinations including vector databases.

Common use cases

  • RAG data preparation
  • Document ingestion for LLMs
  • Knowledge base building
  • Fine-tuning data preparation

Products and services

Unstructured open-source library

Python tools for parsing and chunking documents

Unstructured Platform

Managed pipelines for ingestion, transformation and delivery of unstructured data

Unstructured API

API for converting files into structured JSON

Key features

  • Document partitioning
  • Chunking and enrichment
  • Source and destination connectors
  • Table and layout extraction
  • Metadata extraction
  • Managed workflows

Integrations

  • Amazon S3
  • Google Drive
  • SharePoint
  • Snowflake
  • Databricks
  • Pinecone
  • Weaviate
  • MongoDB

Unstructured alternatives

FAQs

Is there an open-source option?

Yes, the core library is open source and a hosted platform adds managed pipelines.

What file types are handled?

Common business documents including PDF, Word, PowerPoint, HTML, email and images.

Where does the output go?

Connectors deliver processed data to destinations such as vector databases and object storage.

Work at Unstructured? Request updates to this profile. Update this profile

Compare with similar providers

All Data Infrastructure providers