Morphik Core

Morphik Core is a self-hosted multimodal retrieval engine for developers building AI applications around visually rich documents. It searches diagrams, schematics, charts, and datasheets alongside text, so applications can retrieve information that text extraction alone can miss. You can run it on your own server, including through Docker, or use Morphik's hosted service.

Its visual search uses techniques such as ColPali and represents whole document pages to retain visual context. It supports search across PDFs, images, and videos. Developers can combine document ingestion and retrieval with structured information extraction, metadata labeling, classification, and bounding boxes. Knowledge graphs provide another way to explore the stored information.

A Python SDK and REST API let developers use the engine in their applications. The browser console supports file uploads, connections to data sources, search, and chat with documents. Its web interface can also be embedded in another application with custom branding. MCP support connects it to Claude, Open Web UI, and other MCP clients; data integrations include Google Suite, Slack, and Confluence.

For shared deployments, Morphik Core supports multiple tenants, role-based access control, and access rules expressed in natural language. Deployment choices include on-premises infrastructure and cloud environments such as AWS, GCP, and Azure. Support for self-hosted deployments is limited to installation guidance and community help.

The code uses the Business Source License 1.1, which places conditions on production use. Its additional use grant and commercial terms govern self-hosted deployments.