Skip to content
Glossary

Concept

Vector memory

Storing facts as embeddings for semantic retrieval — Niyra's long-term memory layer.

Instead of storing memory as raw text and keyword-searching, vector memory stores embeddings (high-dimensional numerical representations) and retrieves by semantic similarity. 'When does my insurance expire' finds memories about insurance even if you said 'policy renewal'.

Niyra uses pgvector (Postgres extension) with OpenAI text-embedding-3-small for embeddings.

Related

32+

Integrations

OAuth-first

5

Channels

Web, WhatsApp, Telegram, Discord, voice

100+

Native tools

Memory, voice, browser, automation

18

Skills

JIT-loaded per turn

WhatsAppOutbound callsInbox triageCalendar tetrisLong-term memoryAutomation

Keep going

For AI:.md.txt