For the complete BrainAPI documentation index, see llms.txt. A markdown version of any docs page is available by appending .md to its URL. Docs MCP: /docs/mcp.
Introduction
Get started with BrainAPI
BrainAPI is an on-demand memory layer for your LLM applications.
It is a simple API that allows you to store and retrieve data from a memory store.
It is designed to be used with any LLM application, but it is particularly useful for applications that need to store and retrieve data from a memory store, ai assistants for documentation search, chatbots, etc.
Examples
Next.js simple route example in just 4 steps
import OpenAI from "openai";import { LumenBrainDriver } from "@lumenlabs/lumen-brain";const openai = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });const brain = new LumenBrainDriver(process.env.LUMEN_BRAIN_API_KEY!);export async function POST(req: Request) { try { const { message, memoryId, conversationId } = await req.json(); // 1. Save the sent message to the memory ======== const brainResponse = await brain.saveMessage( memoryId, message, "user", conversationId ); // ============================================ // 2. Query the memory for relevant context ====== const queryResponse = await brain.queryMemory( message, brainResponse.memory_id, brainResponse.conversation_id! ); // ============================================ // 3. Send the request to the LLM ================ const response = await openai.responses.create({ model: "gpt-4o", input: message + queryResponse.context, }); // =========================================== // 4. Save the response to the memory ============ await brain.saveMessage( brainResponse.memory_id, response.choices[0].message.content!, "assistant", brainResponse.conversation_id! ); // ============================================ return response; } catch (error) { return error; }}See Next.js example chat with streaming and memory project here