alma-memory is an open-source Python library that provides persistent memory architecture for AI agents. It supports hybrid search, retrieval-augmented generation (RAG), and integrates with Azure and Anthropic models, making it ideal for developers building advanced AI systems.
In the RAG, search & retrieval space, alma-memory takes a focused approach. It focuses on enabling AI agents to store and retrieve persistent memory for improved context and learning. It is built as an open-source project for AI developers. alma-memory is open source under the MIT license. It runs on the command line, and it can be self-hosted.
Behind alma-memory is RBKunnela, and the product first shipped in 2026. Development happens publicly on GitHub with 45 stars and 66 commits in the last 90 days. PulseGate's similarity index finds few close equivalents — alma-memory occupies a relatively distinct niche. Key capabilities include persistent memory, hybrid search, and RAG support.
Latest indexed changes and source events
Other apps tracked under the same category.
