sayoDB CLI (sayodb)

The In-Memory Vector Store & Key-Value Database That Never Runs Out of Memory.

High-performance in-memory database featuring Zero-OOM Tiered Spilling to disk, embedded Cosine Similarity Float32 Vector Engine, and Protocol-Level JSON Schema Validation. Compatible with RESP (Redis socket) and HTTP REST protocols.

Running the sayoDB Engine

Before connecting with the CLI, start a sayoDB database server instance using Docker:

docker run -d --name sayodb -p 6380:6380 -p 6381:6381 sanjoydb/sayodb-server:latest

• Server Docker Image: sanjoydb/sayodb-server:latest
• Web GUI Dashboard: sanjoydb/sayodb-gui:latest (Access at http://localhost:8080)

Instant Execution (No Installation Required)

Run the interactive terminal CLI instantly with npx or bunx:

npx sayodb
# or
bunx sayodb

Installation

Install globally as a CLI tool or as a project dependency across npm, pnpm, yarn, or bun:

As a Project Dependency:

npm install sayodb
# or
pnpm add sayodb
# or
yarn add sayodb
# or
bun add sayodb

As a Global Terminal CLI:

npm install -g sayodb
# or
pnpm add -g sayodb

Quickstart CLI Commands

Once connected to your sayoDB server (default 127.0.0.1:6380):

# Plain-English Key-Value Operations
STORE user:101 "Alice"
FETCH user:101
REMOVE user:101

# AI & Semantic Vector Search Commands
SEMSET "what is sayodb" "In-memory vector store" EMBEDDING 0.82 0.15 0.54
SEMGET THRESHOLD 0.85 EMBEDDING 0.80 0.18 0.52

# Protocol-Level JSON Schema Enforcement
SCHEMA SET user_schema '{"name":"string","age":"number"}'
SETJSON user:101 SCHEMA user_schema '{"name":"Alice","age":25}'

License

MIT © 2026 sayoDB Team