Supermemory gives your LLMs long-term memory. Instead of stateless text generation, they recall the right facts from your files, chats, and tools, so responses stay consistent, contextual, and personal.

​
How does it work? (at a glance)

  • You send Supermemory text, files, and chats.
  • Supermemory intelligently indexes them and builds a semantic understanding graph on top of an entity (e.g., a user, a document, a project, an organization).
  • At query time, we fetch only the most relevant context and pass it to your models.
We offer three ways to add memory to your LLMs:

​
Memory API — full control

  • Ingest text, files, and chats (supports multi-modal); search & filter; re-rank results.
  • Modelled after the actual human brain’s working with smart forgetting, decay, recency bias, context rewriting, etc.
  • API + SDKs for Node & Python; designed to scale in production.
You can reference the full API documentation for the Memory API here.

​
AI SDK

  • Native Vercel AI SDK integration with @supermemory/tools/ai-sdk
  • Memory tools for agents or infinite chat for automatic context
  • Works with streamText, generateText, and all AI SDK features
import { streamText } from "ai"
import { supermemoryTools } from "@supermemory/tools/ai-sdk"

const result = await streamText({
  model: anthropic("claude-3"),
  tools: supermemoryTools("YOUR_KEY")
})
The AI SDK is recommended for new projects using Vercel AI SDK. The Router works best for existing chat applications, whereas the Memory API works as a complete memory database with granular control.

​
Memory Router — drop-in proxy with minimal code

  • Keep your existing LLM client; just append api.supermemory.ai/v3/ to your base URL.
  • Automatic chunking and token management that fits your context window.
  • Adds minimal latency on top of existing LLM requests.
All three approaches share the same memory pool when using the same user ID. You can mix and match based on your needs.

​
Next steps

Head to the Router vs API guide to understand the technical differences between the two and pick what’s best for you with a simple 4-question flow.