Sensible Instruct: Transform documents into structured data with GPT-4

Updated on
October 10, 2023
5
min read
Contributors
No items found.
Author
Sensible Instruct: Transform documents into structured data with GPT-4
Table of contents
Turn documents into structured data
Get started free
Share this post

Today, we're thrilled to launch Sensible Instruct, a document understanding tool powered by LLMs, including GPT-4. With Sensible Instruct it's incredibly easy for anyone, from developers, to data scientists, to business operations managers, to transform documents into useful, structured data.

With Sensible Instruct, you use natural language to instantly extract data from any document, even ones you’ve never seen before. Resumes, invoices, contracts, academic research, bank statements, utility bills and more – Sensible Instruct can parse them all.

Describe your data. Extract your results. 

Simplify your document processing with Sensible Instruct's intuitive and powerful document extraction capabilities:

  • AI-powered data extraction: Leverage GPT-4 to extract accurate and reliable data from your documents.
  • Extract accurate data using natural language: Just use everyday language. It's intuitive and easy to onboard teammates.
  • Parse any document: Extract from a wide range of documents – unstructured and structured.
  • Customizable: Combine Sensible Instruct with SenseML to build tailored configurations.

Extract accurate data from any document using natural language

Sensible Instruct is versatile and can extract data from a wide variety of documents. To achieve this, we’ve introduced three powerful primitives that accept a natural language description of the data you wish to extract:

Query: Extracts individual facts from a document, such as the due date from an invoice, the liability limit from an insurance policy, the address from a passport, or the account number from a bank statement.

List: Extracts repeating data elements from a document, such as the work experience or skills from a resume, a patient's allergies from medical records, or the line items from a purchase order.

Table: Extracts data from tables from a document, such as the transactions from a credit card statement, the balance sheet from a 10-Q regulatory filing, or covered vehicles from an insurance policy.

Bridging the Gap: Sensible Instruct for Developers and Non-Technical Staff

Sensible Instruct's intuitive interface empowers non-technical team members to create and test natural language prompts, reducing the burden on developers. To unlock the full potential of Sensible Instruct, deploy your parsers as APIs to extract tens of thousands of documents an hour, or automate your workflow with our Zapier integration to connect your data to 5,000+ business apps.

Support for other LLMs

As AI advances, we expect the rapid emergence of increasingly powerful and domain-specific LLMs. Sensible Instruct is designed with that in mind: enabling Sensible to support and integrate future state-of-the-art LLMs as they become available. Today we use OpenAI’s GPT models. Tomorrow… 

Sensible Instruct: extract data from documents with LLMs for free

Sensible Instruct is available today at no additional cost during this initial phase. We are actively researching potential use-cases for further development, and your feedback and insights will help shape Sensible Instruct’s future.

Give Sensible Instruct a try to experience its powerful document extraction for free or request a demo.

Turn documents into structured data
Get started free
Share this post

Turn documents into structured data

Stop relying on manual data entry. With Sensible, claim back valuable time, your ops team will thank you, and you can deliver a superior user experience. It’s a win-win.