Unstract: Open-source platform to ship document extraction APIs/MCPs in minutes (github.com)

Unstract has introduced an open-source platform that enables users to develop and deploy document extraction APIs in just minutes, promising close to 100% accuracy for document-centric workflows. The platform features Prompt Studio, a dedicated environment for schema definition that allows users to compare outputs from various large language models (LLMs), track costs, and refine prompts across diverse document types. Once the schema is defined, Unstract simplifies integration into existing systems through various deployment options, including MCP servers, REST API endpoints, ETL pipelines, and low-code nodes for automation. This development is significant for the AI/ML community as it addresses a critical pain point in document processing by enhancing efficiency and reducing costs—key elements given the expansive use of documents in enterprise settings. Noteworthy features include "SinglePass Extraction," which can cut LLM token usage by up to 8x, and the "LLMChallenge," which utilizes dual LLMs to ensure reliability in output. Unstract's extensive support for multiple file formats and third-party integrations makes it a versatile tool for developers and data engineers, further accelerating automation in document-based tasks.