Research and Development|Live

Mem360

Universal AI memory engine that extracts, stores, and retrieves long-term user memories for any AI application. Pluggable LLM, storage, and embedding layers.

Overview

A clean pipeline: conversation > LLM extraction > embedding > vector storage > semantic retrieval > prompt injection. Pluggable at every layer. Works as Python SDK, REST API, or CLI.

Specs

Pipelineconversation > LLM extraction > embedding > vector storage > retrieval

Embeddings384-dim vectors (MiniLM-L6-v2)

DedupCosine similarity >= 0.85 triggers merge/skip

AccessPython SDK, REST API, CLI

Features

mem.add(): LLM extraction > validation > deduplication > embed > upsert
mem.search(): Embed query > vector search > re-ranking (relevance + confidence + recency)
mem.get_context(): Formats results into system prompt preamble
Memory lifecycle: Created > Accessed > Decayed > Archived > Deleted
Pluggable: OpenAI/Anthropic/Groq/Ollama, In-Memory/JSON/SQLite/Qdrant

Tech Stack

PythonFastAPIsentence-transformersQdrantSQLiteasyncio

Interested in Mem360?

Let's discuss how this can work for you.

Get in Touch