Morphik
Morphik Core is an open-source multimodal retrieval engine that helps developers integrate complex, visually rich documents into AI applications. It solves the problem of fragmented RAG pipelines by providing a unified toolset for ingesting, searching, and managing unstructured data. It is for developers building AI applications that need to understand charts, diagrams, tables, and other visual content.
✨ Key features
- Multimodal search using ColPali for visual content understanding
- Fast and scalable metadata extraction including bounding boxes and classification
- Integrations with Google Suite, Slack, and Confluence
- Python SDK and REST API for easy integration
- Web-based Morphik Console for interactive data management
- Model Context Protocol (MCP) support
🎯 Use cases
- Build AI applications that answer questions from technical manuals with diagrams
- Search and retrieve information from scanned PDFs and images
- Automate back-office operations like AP, billing, and payroll with document understanding
- Create chat interfaces that query complex multimodal documents
📦 Installation
🧰 Requirements: Requires a Morphik Core instance (cloud or self-hosted) and an API key or URI. Python SDK requires Python.
For the recommended cloud setup, sign up at dev.morphik.ai and use the Python SDK. For self-hosting, follow the dedicated installation guide at https://dev.morphik.ai/docs/getting-started (offers direct and Docker installation).
🚀 Usage
from morphik import Morphik
morphik = Morphik("<your-morphik-uri>")
morphik.ingest_file("path/to/your/super/complex/file.pdf")
# Query your data
morphik.query("What's the height of screw 14-A in the chair assembly instructions?")
⚠️ Good to know
Self-hosted deployments are not fully supported due to limited resources; only an installation guide and community Discord are provided.
❓ FAQ
How do I get started with Morphik Core?
The recommended way is to sign up for a free account at dev.morphik.ai and use the Python SDK or REST API. You can also self-host by following the installation guide.
What types of documents can Morphik Core handle?
It handles unstructured and multimodal documents including PDFs, images, videos, and more, with search that understands visual content.
Is Morphik Core free to use?
It has a free tier for the cloud service. For self-hosted, it's free for personal/indie use and for commercial use if your deployment generates less than $2,000/month in gross revenue; otherwise you need a commercial key.
Can I contribute to Morphik Core?
Yes, contributions are welcome via GitHub issues and pull requests. They are currently focused on improving speed, integrations, and research papers.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.