Integrate persistent memory into your product. Store, classify, connect, and retrieve information via API — with enterprise-grade security and compliance.
Developer API
12 RESTful endpoints with API key authentication. Store, query, and manage memories programmatically.
Technical Features
Every component built for production workloads with full compliance and audit capabilities.
7 entity types (Person, Place, Organization, Product, Event, Concept, Date) and 11 relationship types extracted automatically from every record.
pgvector + GIN indexes4-layer search: Entity Graph lookup, Semantic (pgvector cosine similarity), Lexical (full-text + trigram), and Temporal filters combined via Reciprocal Rank Fusion.
RRF scoringAutomatic categorization across 9 categories with sentiment analysis, confidence scoring, and entity extraction. Powered by Claude (Anthropic).
Claude 3.5PostgreSQL RLS policies ensure complete tenant isolation. Each API key is scoped to a single user's data. Zero cross-tenant data leakage.
PostgreSQL RLSRecords stored verbatim and never modified. Corrections create new versions with full audit trail. Original data preserved with timestamps.
Append-onlyModel Context Protocol support coming soon. Connect Memorize directly to LLM agents as a persistent memory layer for AI applications.
Coming soonUse Cases
Patient history, allergies, medications, and preferences. Retrieve critical information instantly during consultations.
Case notes, client details, legal precedents. Immutable records with full audit trail for compliance requirements.
Client preferences, meeting notes, deal context. Knowledge graph connects people, companies, and interactions automatically.
Interaction memory across channels. Agents instantly recall past conversations, preferences, and resolutions.
Study notes, lecture summaries, research references. Students and researchers find exactly what they captured.
Security & Compliance
Full compliance with Brazilian and European data protection laws. Data export (Article 20) and right to erasure built-in.
Sensitive fields encrypted at rest with per-user keys. Crypto-shredding enables complete data erasure by destroying the key.
Complete log of all operations — creation, access, corrections, deletions. Immutable records ensure regulatory compliance.
PostgreSQL Row-Level Security ensures zero cross-tenant data access. Each user's data is completely isolated at the database level.
Pricing
For individual developers
For teams and production workloads
FAQ
Each user gets a unique API key. Include it in the Authorization header as a Bearer token. Keys are scoped to a single user and can be regenerated at any time.
All endpoints accept and return JSON. Text content is stored verbatim. Audio can be sent via Telegram/WhatsApp and is transcribed automatically via Whisper AI.
The free tier has 10 queries per day. Enterprise plans offer unlimited queries with custom rate limits based on your needs. Contact sales for volume pricing.
PostgreSQL Row-Level Security (RLS) policies ensure each API key can only access its own user's data. This is enforced at the database level, making cross-tenant access impossible.
Yes. The /export endpoint provides full GDPR-compliant data export. For deletion, crypto-shredding destroys the encryption key making all data permanently unrecoverable.
Model Context Protocol (MCP) support is on our roadmap. It will allow LLM agents to use Memorize as a persistent memory layer, enabling AI applications with long-term context.
Contact
Want to integrate Memorize into your product, need custom volumes, or have questions about the API? Our team is here to help.
Add persistent, intelligent memory to your application. Free to start, scales with you.
Explore the API →