Sold by: Unstructured
Deployed on AWS
AWS Free Tier
Unstructured extracts and transforms data for use with every major vector database and LLM framework
5
Overview
Ingest and preprocess complex natural language data from any document, file type, or layout with Unstructured.
Under the hood, the Unstructured engine involves breaking a document into its constituent parts and identifying the document's structure, such as its header, tables, and body text. Unstructured provides diverse preprocessing strategies for documents each catering to different document types and requirements. Utilizing the optimal strategy enhances document element classification accuracy and extraction efficiency, which is crucial for image-based files and layout-intensive documents.
Click on Continue to Subscribe to start using Unstructured for your data preprocessing needs.
We are constantly improving our products and love feedback.
Highlights
- Transforms all your data for downstream analytics. Next-generation vision transformer for images, PDF, and table extraction
- Enhanced models for table extraction, document hierarchy, and element classification. Chunks your data for LLM applications
- Compatible with any embedding model, vector database, and LLM framework. API client libraries in multiple client languages (e.g. Python, Javascript)
Details
Sold by
Delivery method
Delivery option
UnstructuredAPI
64-bit (x86) Amazon Machine Image (AMI)
Latest version
Operating system
OtherLinux 9
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Trust Center
Access real-time vendor security and compliance information through their Trust Center powered by Drata or Vanta. Review certifications and security standards before purchase.
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
If you are an AWS Free Tier customer with a free plan, you are eligible to subscribe to this offer. You can use free credits to cover the cost of eligible AWS infrastructure. See AWS Free Tier for more details. If you created an AWS account before July 15th, 2025, and qualify for the Legacy AWS Free Tier, Amazon EC2 charges for Micro instances are free for up to 750 hours per month. See Legacy AWS Free Tier for more details.
Dimension | Cost/hour |
|---|---|
m4.xlarge Recommended | $4.40 |
t2.micro | $1.10 |
t3.micro | $2.20 |
m4.4xlarge | $17.60 |
r5.24xlarge | $105.60 |
r5.large | $2.20 |
r5.8xlarge | $35.20 |
m5n.16xlarge | $70.40 |
r3.xlarge | $4.40 |
i3.2xlarge | $8.80 |
Vendor refund policy
We do not currently support refunds, but you can cancel at any time.
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
64-bit (x86) Amazon Machine Image (AMI)
Amazon Machine Image (AMI)
An AMI is a virtual image that provides the information required to launch an instance. Amazon EC2 (Elastic Compute Cloud) instances are virtual servers on which you can run your applications and workloads, offering varying combinations of CPU, memory, storage, and networking resources. You can launch as many instances from as many different AMIs as you need.
Version release notes
1.0.72
- update contextual chunking chunk window size to be 100k from 15k
Additional details
Usage instructions
"To connect to the operating system, use SSH and the username rocky. You will need the same SSH Key Pair supplied during stack launch. For more details see the Unstructed API deployment guide here: https://docs.unstructured.io/api-reference/api-services/aws "
Resources
Vendor resources
Support
Vendor support
Please allow 24 hours. Join us in our Slack workspace for support
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Updated weekly
By Unstructured
By Hyperscience
By SS&C Blue Prism
Top
10
In Software Development, Data Analysis
Top
10
In Text/OCR
Top
10
In Handwriting Recognition
Sentiment is AI generated from actual customer reviews on AWS and G2
Functionality
Ease of use
Customer service
Cost effectiveness
Positive reviews
Mixed reviews
Negative reviews
AI generated from product descriptions
Document Structure Analysis
Breaks documents into constituent parts and identifies structural elements including headers, tables, and body text
Vision Transformer Technology
Utilizes next-generation vision transformer models for extraction from images, PDFs, and tables
Multi-Format Data Ingestion
Ingests and preprocesses complex natural language data from any document type and file layout
Vector Database and LLM Integration
Compatible with any embedding model, vector database, and LLM framework with API client libraries in multiple languages including Python and Javascript
Adaptive Preprocessing Strategies
Provides diverse preprocessing strategies tailored to different document types with enhanced models for table extraction, document hierarchy, and element classification
Production-Grade Runtime Environment
Pre-built and optimized Python runtime for document processing tasks with rapid deployment capabilities
Zero-Shot Document Templates
Accelerated processing blueprints for common document types including forms with handwriting, paystubs, bank statements, and invoices without requiring training data
Comprehensive Document Intelligence
Built-in capabilities for document classification, extraction, transcription, complex table handling, and Human-in-the-Loop interfaces with native QA support
High-Throughput Processing Architecture
Massively parallel processing engineered for high-throughput, low-latency performance at scale with intelligent cost orchestration across CPU and GPU resources
Model-Agnostic Pipeline Framework
Support for any model type or custom third-party integration within a single pipeline with centralized versioning, auditability, and full telemetry tracking on all inputs, outputs, and automated decisions
Handwritten and Printed Text Recognition
Achieves 98% accuracy in extracting printed and handwritten text, numbers, and checkboxes from low-quality faxed and scanned documents using machine learning trained on over 1 billion documents from finance, insurance, and healthcare industries.
Document Classification and Matching
Automatically classifies and identifies pages by matching them to user-defined form templates using image recognition algorithms, capable of handling degraded documents including those with stains, incorrect orientation, and poor quality scans.
Signature and Barcode Detection
Detects signatures, QR codes, barcodes, and performs address validation with specialized readers available for advanced use cases.
Template-Based Form Field Extraction
Uses blank forms as templates to locate and extract individual form fields for digitization without requiring coding or machine learning expertise through a web-based developer interface.
Data Transformation and Human-in-the-Loop Review
Applies data transformations to ensure output consistency and readiness for RPA ingestion, with optional human-in-the-loop review interface for data validation and business logic application, integrated via API or Blue Prism connector.
Standard contract
No
No
Customer reviews
reviewer2846073
Fast document parsing has boosted culture insights and now improves HR policy intelligence
Reviewed on Jun 04, 2026
Review from a verified AWS customer
What is our primary use case?
We are a culture operating system that analyzes organizational culture, and we have an AI bot that joins calls to create structured culture intelligence reports. When we talk about HR, there are HR policies, PDFs, and performance documents generated by the HR or human resource department in the company. If we need to digest that data, we use Unstructured to create a vector database of these unstructured data.
If an HR manager wants to use HR policies, HR documents, and performance data in Instill, they can upload their document, and we use Unstructured to convert those PDFs into a vector-based database.
What is most valuable?
We are now using Unstructured every day, and it is useful when we want answers and AI to be used on a PDF or something similar. We use Unstructured to convert it into a vector database to make retrieval augmentation or any kind of AI processes easy.
The document parsing stands out, as the document ingestion is very fast in Unstructured, 20 to 40% faster than the industry products available. If HR wants to upload a PDF on our platform, we use Unstructured to digest the data, and it is 20 to 40% faster than other solutions.
The faster document ingestion has resulted in customer satisfaction, leading to higher quality answers using AI that improved customer satisfaction and NPS score. NPS has improved by at least 10 to 15 points since we started using Unstructured, not only for data digestion but also for retrieving data when we have to use AI or RAG.
What needs improvement?
Cost is something that needs to be factored for scaling use cases because we do not have control over how many documents users will upload, so it is variable and we cannot set a threshold.
For how long have I used the solution?
I have been working in my field for the last eight years.
What do I think about the stability of the solution?
The accuracy and reliability of output from Unstructured are very accurate and highly reliable, as we have not faced any issues and the uptime is consistent.
Which other solutions did I evaluate?
I advise doing research about other vector database searches because Pinecone is also good, but you need to understand the use case.
What other advice do I have?
Features and usability are fine, and it is one of the best products available.
I chose a rating of 10 out of 10 because they are very focused on doing what they do at the best quality and speed, and what they are not doing is outside their scope. They claim faster processing and converting into a vector database faster, building a vector database from unstructured data, which they provide at a very fast speed and quality.
The governance and security regarding Unstructured's AI capabilities are good, as we have SOC 2 and other compliance certificates from Unstructured. I give this product a rating of 10 out of 10.
Which deployment model are you using for this solution?
Private Cloud