Overview

Product video
LLMWhisperer is a document parser that prepares documents for large language models.
It takes PDFs, scans, images, and just about any documents. It then returns text that preserves the layout of the source, so your LLMs receive maximum context.
It works in three parts: text extraction, document handling, and integration.
Text Extraction LLMWhisperer reads a document and returns text that preserves its original structure, ready to pass to any model.
- Preserves multi-column layouts, tables, and key-value pairs in place.
- Returns tagged, layout-preserved output rather than a flat block of text.
- Reaches 99.9% accurate raw text extraction across supported document types.
Document Handling LLMWhisperer processes documents across various formats, with no manual pre-processing.
- Reads scans, image PDFs, rotated pages, and handwriting as they come.
- Extracts across 300+ languages, including non-Latin scripts.
Integration LLMWhisperer runs as a REST API you call directly, built to drop into existing deployments or pipelines.
- Extracts in synchronous or asynchronous mode, with Webhooks that signal when a job finishes.
- Ships with Python and Node.js clients to shorten integration.
- Processes pages in parallel to reduce manual work and save time.
Common use cases
- Document pre-processing
- Feeding scanned and image PDFs into LLM workflows
- Converting handwritten forms into machine-readable text
- Preparing invoices, contracts, and statements for downstream extraction
- Building searchable text from scanned archives
- Multilingual document processing across regions
Highlights
- Layout-preserving mode: keeps multi-column text, tables, and forms in their original structure, so models read a document the way a person would.
- 300+ language coverage: extracts mixed-language and non-Latin documents, including Arabic, Chinese, Japanese, and Hindi.
- Highlight API: returns the location of every extracted line in the source document, so you can highlight where a value came from during human review.
Details
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
Pricing
Dimension | Description | Cost/12 months |
|---|---|---|
300k pages | $36,000 per year to process 300,000 pages per year. | $36,000.00 |
The following dimensions are not included in the contract terms, which will be charged based on your usage.
Dimension | Cost/unit |
|---|---|
Overage @ $0.05 per page. | $0.05 |
Vendor refund policy
This is a contract with usage-based pricing. Once the usage credits included in the contract are exhausted, you will be billed based on your actual usage.
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
Software as a Service (SaaS)
SaaS delivers cloud-based software applications directly to customers over the internet. You can access these applications through a subscription model. You will pay recurring monthly usage fees through your AWS bill, while AWS handles deployment and infrastructure management, ensuring scalability, reliability, and seamless integration with other AWS services.
Resources
Vendor resources
Support
Vendor support
If you need assistance or encounter any issues during use, please contact us at support@unstract.com .
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.