12+ engines • SQL, document, cache, vector & ledger
100% Free & Open Source

Your databases, in one local-first workspace

KamehaDB is a cross-platform desktop GUI for SQL, document, cache, vector, and ledger systems — built with AI in, not bolted on. Runs entirely on your machine.

11 Stars
macOS
Windows
Linux
<query.sql
kamehadb@local:~$
12+ engines — SQL, document, cache, vector & ledger

Free. Local. Yours.

Four principles. No asterisks.

100% Free & Open Source

No subscriptions, no premium tiers, no telemetry. The full source is on GitHub under Apache-2.0.

Local-First & Private

Connections, credentials, query history, and chat logs stay on your machine. We have no servers — there is nowhere for your data to go.

Bring Your Own AI

OpenAI, Ollama (local + cloud), 9router — pick the model that fits your workflow, or skip AI entirely.

OpenAIOllama LocalOllama Cloud9router

Open Standards

Pure SQL, MongoDB aggregation pipelines, Redis protocol. No proprietary query language, no migration lock-in.

Just say what you need

Describe your intent in plain English. KamehaDB writes the SQL for you.

You asks
KamehaDB writes

Everything you need

Powerful features to streamline your database workflow

One workspace, 12 engines

Relational, document, cache, vector, and ledger systems in a single desktop app. Browse schemas, collections, keyspaces, points, and accounts side by side.

AI that knows your schema

Describe what you want in plain English. The assistant reads your real DDL, indexes, and constraints to generate SQL, MongoDB pipelines, Redis commands, and Qdrant searches.

Global search — Ctrl+K

Fuzzy-find any connection, table, column, or open tab. Jump into a query, a stats page, or the AI chat without leaving the keyboard.

Contextual AI chat

Persistent history per connection, streamed responses, and a Run button to execute the SQL the assistant writes. Bring your own OpenAI, Ollama, or 9Router.

Schema timeline, diff & migration

Capture snapshots on demand, compare schemas side by side with per-table change cards, and generate the DDL to migrate from one state to the next — without writing ALTER TABLE by hand.

Built-in charts

Turn any query result or Mongo collection into bar, line, area, or pie charts. Histograms, aggregates, and time series without exporting to a notebook.

Query history with favorites

Normalized query history per connection, grouped by pattern with duration stats. Star the ones you reach for every day; reuse, fork, and rerun in one click.

Engine-native tooling

PostgreSQL stats, MongoDB explorer with chart view, Redis key browser, Qdrant 3D vector map, TigerBeetle accounts/transfers — purpose-built screens for each engine.

Get started in minutes

Install KamehaDB and connect to your first database in under 2 minutes. Works on macOS, Windows, and Linux.

1

Download the app

Available for macOS, Windows, and Linux

2

Add your connection

PostgreSQL, MySQL, MariaDB, SQLite, SQL Server, Oracle, ClickHouse, DuckDB, MongoDB, Redis, Qdrant, or TigerBeetle

3

Start exploring

Browse schemas, run queries, and chat with AI

# Clone the repository
git clone https://github.com/asta-nguyen/kamehadb.git
# Install dependencies and run
cd kamehadb
pnpm install && pnpm dev
# Build you app in your OS
pnpm tauri build --target aarch64-apple-darwin
# Build you app in your Windows
pnpm tauri build --target x86_64-pc-windows-msvc
# Build you app in your Linux
pnpm tauri build --target x86_64-unknown-linux-gnu

One app for every database you run

Free, open source, and local-first. No telemetry, no cloud proxy — your data stays on your machine.

Download for free