RAGFlow
Self-Hosted AIRAG engine with deep document understanding
Replaces NotebookLM
The 12 best PostgreSQL self-hosted ai you can self-host, ranked by community traction.
RAG engine with deep document understanding
Replaces NotebookLM
Self-hosted memory layer for AI agents and assistants
Replaces OpenAI Memory
Open-source observability and analytics for LLM apps
Replaces LangSmith
Memory and knowledge graph engine for AI agents
Replaces Mem0, Pinecone
Build stateful LLM agents with long-term memory
Replaces OpenAI Assistants
Knowledge base question-answering system built on LLMs
Replaces Intercom Fin
Open-source GenBI agent for talking to your database
Replaces ThoughtSpot
Open-source framework for autonomous AI agents
Replaces AutoGPT
Customizable AI research agent connected to your knowledge
Replaces NotebookLM, Perplexity
Open LLM application platform for enterprise workflows
Replaces Dify
Production-ready RAG engine with a RESTful API
Replaces OpenAI Assistants
On-premise generative AI platform for teams
Replaces ChatGPT Enterprise
No apps match these filters.
| App | Deploy effort | Difficulty | Min. RAM | License | ★ Stars |
|---|---|---|---|---|---|
| RAGFlow | Under-an-hour setup | Medium | 8,192 MB | Apache-2.0 | 92,165 |
| Mem0 | Under-an-hour setup | Medium | 1,024 MB | Apache-2.0 | 67,140 |
| Langfuse | Under-an-hour setup | Medium | 2,048 MB | MIT | 35,795 |
| Cognee | Under-an-hour setup | Medium | 2,048 MB | Apache-2.0 | 32,226 |
| Letta | Under-an-hour setup | Medium | 2,048 MB | Apache-2.0 | 25,125 |
| MaxKB | Under-an-hour setup | Easy | 2,048 MB | GPL-3.0 | 22,973 |
| Wren AI | Under-an-hour setup | Medium | 4,096 MB | AGPL-3.0 | 17,865 |
| SuperAGI | Under-an-hour setup | Hard | 4,096 MB | MIT | 17,702 |
| SurfSense | Under-an-hour setup | Medium | 2,048 MB | Apache-2.0 | 16,423 |
| Bisheng | Under-an-hour setup | Hard | 4,096 MB | Apache-2.0 | 12,044 |
Every option here is open-source and self-hostable, tagged PostgreSQL within the self-hosted ai category. Deploy effort is graded from published container recipes, not vendor claims — see the methodology.