Verified project record

miurla/morphic

morphic is an AI-powered search engine that delivers grounded, cited answers through a rich generative UI. It supports multiple configurable search and language model providers, enabling flexible and self-hosted search experiences.

Project overview

It distinguishes itself by rendering live, rich inline components from a streamed JSON spec rather than plain markdown, and its Docker Compose setup automatically bundles SearXNG, eliminating the strict need for external search API keys.

Project type: RAG · AI Search · Infrastructure
Use cases: Search & Research
Deployment: Refer to project documentation
License: Apache-2.0

Best for

General users and developers seeking a self-hostable, AI-driven search application that provides cited answers with a rich user interface.

Key capabilities

Provides AI-powered search returning grounded, cited answers to user queries.
Renders rich inline components (source-credited images, grids, headings) live from a streamed JSON spec, beyond plain markdown.
Offers Quick and Adaptive search modes.
Dynamically detects and allows switching between OpenAI, Anthropic, Google, Ollama, Vercel AI Gateway, and OpenAI-compatible providers.
Supports Tavily, SearXNG, Brave, and Exa as search providers.
Stores user chat history in PostgreSQL.
Enables sharing search results using unique URLs.
Allows users to upload files to the application.

Limitations and risks

The application requires the configuration of at least one AI provider API key (such as OpenAI or Anthropic) or a local model setup via Ollama to function.

Getting started

Run `docker pull ghcr.io/miurla/morphic:latest`, clone the repository to edit `.env.local`, execute `docker compose up -d`, and visit `http://localhost:3000` to access the application.

Evidence and sources

GitHub project description: An AI-powered search engine with a generative UI
README: An AI-powered search engine with a generative UI.
README: - AI-powered search with grounded, cited answers - Generative UI — answers render rich inline components (source-credited images, grids, headings) live from a streamed JSON spec,…
README: The quickest way to run Morphic locally: ```bash docker pull ghcr.io/miurla/morphic:latest ``` Then set up with Docker Compose: 1. Clone the repository and configure environment:…
README: 1. Clone and install: ```bash git clone https://github.com/miurla/morphic.git cd morphic bun install ``` 2. Configure environment: ```bash cp .env.local.example .env.local ``` Edi…

Reviewed at: 2026-07-18T04:08:22Z