jkanalakis/deep-recall
Enterprise-grade memory framework for LLMs featuring GPU-optimized inference, vector storage, and automated scaling. Enables hyper-personalized responses through efficient context retrieval and integration.
Builds on PostgreSQL with pgvector for vector storage and supports pluggable backends (FAISS, Qdrant, Milvus, Chroma) for semantic search across user interaction history. Uses a three-tier microservices architecture—Memory Service for embeddings/retrieval, Inference Service for GPU-optimized LLM inference, and an Orchestrator API gateway—deployed via Docker Compose with Kubernetes support. Includes privacy-first APIs for viewing, updating, or deleting user memories, plus comprehensive monitoring and test suites across unit, integration, and API layers.
No commits in the last 6 months.
Stars
90
Forks
12
Language
Python
License
—
Category
Last pushed
May 03, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/jkanalakis/deep-recall"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
agiresearch/A-mem
A-MEM: Agentic Memory for LLM Agents
gschaidergabriel/lcme
Neural-enhanced conversational memory for AI agents — 10 micro-networks, tri-hybrid storage,...
garan0613/ai-memory-gateway
🧠 给AI加长期记忆的轻量网关 / A lightweight gateway that adds long-term memory to any LLM
cerebralos-org/cerebralos
The Layer of Subconsciousness for AI Agents. Built to forget. Designed to dream.
Djsand/neverforget
LLMs have amnesia. NeverForget is the cure. A universal, infinite-memory proxy for any LLM API.