Memor offers a memory subsystem for language model agents, allowing them to store embeddings of past events, user preferences, and contextual data in vector databases. It supports multiple backends such as FAISS, ElasticSearch, and in-memory stores. Using semantic similarity search, agents can retrieve relevant memories based on query embeddings and metadata filters. Memor’s customizable memory pipelines include chunking, indexing, and eviction policies, ensuring scalable, long-term context management. Integrate it within your agent’s workflow to enrich prompts with dynamic historical context and boost response relevance over multi-session interactions.

RecurSearch is an open-source Python library designed to improve Retrieval-Augmented Generation (RAG) and AI agent workflows by enabling recursive semantic search. Users define a search pipeline that embeds queries and documents into vector spaces, then iteratively refines queries based on prior results, applies metadata or keyword filters, and summarizes or aggregates findings. This step-by-step refinement yields higher precision, reduces API calls, and helps agents surface deeply nested or context-specific information from large corpora.