Welcome to the Mpalo API

Integrate experience-aware memory infrastructure, including episodic recall and context assembly, into your applications using Palo Nano, Palo Bloom, and Palo Research. Explore documentation, guides, and tools to build systems that remember with context, not just retrieve by similarity.

Quickstart → Explore Palo Bloom

Your Data, Your Control

Mpalo is built with privacy and security as foundational principles. We do not train shared base models on your private data or memories without your explicit consent. All user data is encrypted in transit (TLS/SSL) and at rest (AES-256). You retain full ownership of your inputs and the memories Palo generates. With Palo Bloom, you have granular control over your memories -- view, delete, and manage what Palo remembers. For enhanced data governance, Mpalo supports a Bring Your Own Vector Store (BYOVS) model, keeping memory data within your designated environment.

Memory Control & Private Memory Spaces

We ensure strict separation of user experiences and memories. Whether you use Mpalo's secure managed storage or your own vector store via BYOVS, you maintain full ownership and control:

Complete user ownership of all memory data.
Independent control over memory creation, retention, and deletion.
No mixing of user experiences or memories between accounts.
Secure, isolated memory storage per user.
Private Memory Spaces (Beta) -- scalable managed memory for deploying Palo to your customers while maintaining strict privacy. Available on Pro and Enterprise plans.

For comprehensive details, see our Trust & Ethics page .