Unstract
A no-code document extraction platform that turns PDFs and scans into JSON. Self-host with Docker and Ollama, or use managed cloud hosting.
Unstract is a self-hosted, no-code platform for teams that need structured data from documents whose layouts vary. It uses LLMs to extract JSON from PDFs, images, scans and office files without building a template for each vendor or form. Its use cases include finance, insurance, healthcare and KYC document processing.
Prompt Studio lets teams describe the fields they want to extract in natural language. They can configure model, embedding, vector database and text extraction adapters for their document-processing workflow.
Extraction workflows can serve existing applications through a REST API or feed databases through ETL pipelines. Sources include Amazon S3, MinIO, Google Drive and SFTP; destinations include Snowflake, BigQuery and PostgreSQL. An n8n node connects extraction to automation workflows, while an MCP server makes it available to AI agents.
The open-source platform uses the AGPL-3.0 license and runs through Docker on Linux or macOS, including Intel and Apple Silicon Macs, with at least 8 GB of RAM. It supports Ollama for local models, OpenAI-compatible APIs and cloud providers such as Anthropic Claude, Azure OpenAI and AWS Bedrock. Managed cloud hosting is also available. Self-hosting runs the platform on your hardware; choosing cloud model or text extraction services sends those processing tasks to external providers. Usage analytics can be disabled.
