Cortex Open Source: Agent Memory Layer
reneil1337 · reddit · 2026-07-16
The author released Cortex: a knowledge graph/memory layer designed for teams and communities, aiming to enable agents to efficiently query locally organized data. The project is open-source under Apache 2.0 and emphasizes running the entire stack on consumer-grade hardware for a maximally private knowledge base.
The post also recommends model combinations: using lower-barrier models for agentic retrieval, visual data ingestion, and embedding. It mentions that cloud models can be integrated via standardized /v1 endpoints, or scaled locally using configurations like DGX Spark or 3090/4090 GPUs. Structurally, the product is divided into Manage and Explore, supporting document uploads, URL scraping, knowledge graph construction, syncing updates with documents, alongside Git integration and a REST API for easy integration with other software.
More from coding & agent
- Codex helps build Valdiluce, an open-world game with climbing, gliding and gondolas — Dimillian · 2026-07-22
- HeyGen adds a media-sourcing skill for coding agents with 75k images and 10k tracks — HeyGen · 2026-07-22
- Agent search bottlenecks are now about variance, not raw latency — rohanpaul_ai · 2026-07-22
- LangSmith adds tracing for Pipecat, LiveKit, OpenAI Realtime, and Gemini Live — LangChain · 2026-07-22
- An MCP server signs every AI agent tool call into a verifiable Merkle chain — Funky_Chicken_22 · 2026-07-22
- Annotated transcript of a Claude Code team interview is now available — trq212 · 2026-07-22