Open Mind

Open source self deployable knowledge base with AI powered search and chat.

Names

• brainhub
• openmind
• openbrain
• newbrain
• anotherbrain
• thirdbrain
• brainx
• memory
• openmind.site

keywords

• mind
• brain
• knowledge
• memory
• info
• doc, link, too narrow
• ama
• open
• base
• hub
• x
• hunt

What problem does this project solve?

• I want to have a knowledge base that I can search with AI
• I want to self deploy the knowledge base

competitors

https://www.privategpt.io/

Stack

Auth: Clerk App logic: next.js VectorDB: Pinecone LLM Orchestration: Langchain.js Models + inference: Replicate Deployment: Fly

chatbase/chatpdf

Operations:

1. Init a nextjs app https://nextjs.org/docs/getting-started/installation

2. Signup and create app on https://clerk.com. Steps to use in nextjs: https://clerk.com/docs/nextjs/get-started-with-nextjs

3. Signup on fly and deploy app: https://fly.io/docs/languages-and-frameworks/nextjs/

4. Install langchain.js: https://js.langchain.com/docs/getting-started/install

5. https://js.langchain.com/docs/modules/indexes/vector_stores/integrations/pinecone

6. https://js.langchain.com/docs/modules/models/llms/integrations#replicate

How does the chat feature work

1. Get content from a link (puppeteer)
2. Split content to smaller chunks if necessary
3. Compute the embeddings of the content
4. Store the embeddings in Pinecone

Q&A example: https://js.langchain.com/docs/modules/chains/index_related_chains/retrieval_qa