Kotaemon

Kotaemon is an open-source RAG user interface enables document question answering with customizable pipelines.

Kotaemon is a clean and customizable retrieval-augmented generation user interface designed for both end users and developers. It allows users to perform question answering on documents with multi-modal support for figures and tables. The system connects to local language models via Ollama and Llama-cpp-python, as well as API providers like OpenAI, Azure, Cohere, Groq, and others. It features hybrid retrieval, re-ranking, advanced citations with PDF previews, and complex reasoning methods like question decomposition and agents. Developers can build custom pipelines using Gradio, while end users benefit from multi-user login, private or public collections, and a configurable settings interface.

Kotaemon at a glance

Best for: Researcher, Software engineer, Founder
Runs on: Web, Linux, macOS, Windows, self-hosted, open source
Pricing: Free
Released: March 2024
Category: AI research tool
Website: github.com/Cinnamon/kotaemon

Key features

Host document QA web user interfaces
Support local and API-based models
Run hybrid retrieval and re-ranking
Parse multi-modal documents with figures and tables
View advanced citations with PDF previews
Execute complex reasoning with agents
Configure settings and prompts via user interface

Use it for

Chatting with private documents
Running local document question answering
Building custom retrieval-augmented generation pipelines
Parsing multi-modal documents with OCR and tables

Alternatives to Kotaemon

WeKnora — Combines RAG, multi-step agents, and automatic Wiki generation in one platform.
My RAG — Compare Python and Java implementations sharing identical data.
Jina — It provides reliable reader and search APIs optimized for LLMs and RAG systems.
RAG-Anything — It unifies text and multimodal document retrieval within a single framework.
LightRAG — It uses a dual-layer architecture to reduce LLM calls and retrieval latency.
UltraRAG — Build complex RAG workflows using modular MCP components and YAML configuration.
RAGFlow — Provides an integrated platform combining data processing with agent workflows.
Decido — Runs locally on CPU while beating published benchmark scores.
RAG Evaluation Starter — Runs five metrics locally with no SaaS signup or mandatory API key required.
Anime Text Dataset — Provides structured anime data with field-level provenance and question-answering pairs.
Orama — Runs in browsers, servers, and edge networks with zero dependencies.
GraphRAG — Enhances LLM reasoning over private data using knowledge graph structures.

All Kotaemon alternatives