Tablething – Free DB Client with AI Chat

TL;DR

TL;DR: Tablething connects to 25+ databases and lets you query them via AI chat, a typed query editor, or inline data editing — all locally, no cloud required.

Source and Accuracy Notes

⚠️ This section is MANDATORY. All links must be verified from actual source, not guessed.

• Project page: tablething.com ← visited and verified
• Source repository: github.com/tablething/tablething ← README read in full
• License: unspecified in README; binary is free to download
• HN launch thread: news.ycombinator.com/item?id=48577807

What Is Tablething?

Tablething is a free, cross-platform database client that ships as a desktop app for macOS, Windows, and Linux. It connects to 25+ data sources — Postgres, MySQL, MongoDB, Redis, SQL Server, SQLite, Elasticsearch, BigQuery, ClickHouse, DynamoDB, Snowflake, Cassandra, Oracle, Databricks, and more — and lets you query them through an AI chat interface, a typed query editor, or inline grid editing.

The tool is local-first: credentials are stored on your machine, data does not leave your network unless you explicitly use a cloud-connected AI provider.

Key Features

AI chat interface. Ask questions in plain English and Tablething generates and runs a query against the connected database. Supports OpenAI, Anthropic, Google Gemini, Groq, OpenAI-compatible APIs, and Ollama. You bring your own API key.

Multi-datasource connections. Create multiple simultaneous connections across different database engines and query them side by side.

Query editor with syntax highlighting. Write and save queries with full autocompletion for your schema.

Inline data editing. Browse, filter, and update records directly in the grid view.

Chart visualization. Render query results as charts without exporting to another tool.

Local-first architecture. Nothing is sent to Tablething’s servers. AI requests go directly to your configured provider API.

Setup Workflow

Step 1: Download and Install

Download the latest release for your platform from the releases page. The app auto-updates once installed.

Supported platforms:

• macOS
• Windows
• Linux

Step 2: Connect a Database

Launch Tablething and click New Connection. Select your database type and fill in the connection details:

Connection name:  local-postgres
Database type:    Postgres
Host:             localhost
Port:             5432
Database:         myapp
Username:         postgres
Password:         [stored locally]

Repeat for any additional data sources you want to work with.

Step 3: Configure AI Provider

Open Settings → AI Providers and add your API key. Supported providers:

• OpenAI
• Anthropic (Claude)
• Google Gemini
• Groq
• Ollama (self-hosted)
• Any OpenAI-compatible API endpoint

Step 4: Query via Chat or Editor

Switch to the AI Chat tab and ask a question:

Show me the top 10 users by total order value

Tablething generates the SQL, shows you the query for review, and executes it on your connected database.

Or use the Query Editor tab for manual SQL with syntax highlighting and saved queries.

Practical Evaluation Checklist

• [ ] Connects to a local Postgres/MySQL instance without errors
• [ ] AI chat generates correct SQL for a simple aggregation query
• [ ] Schema autocompletion works in the query editor
• [ ] Inline edit + save updates a row correctly
• [ ] Chart renders from a SELECT result
• [ ] Second simultaneous connection works without conflict
• [ ] No data sent to Tablething’s servers (verify via network tab)

Security Notes

• Credentials are stored locally on your machine.
• AI queries are sent directly to your configured provider (OpenAI, Anthropic, etc.) — not through a Tablething proxy.
• No telemetry or analytics collection reported in the README.
• Swarm secrets storage is used for registry credentials when relevant.

FAQ

Q: Does Tablething work offline? A: Yes. The app and all local connections work fully offline. AI chat requires network access only to reach your configured provider API.

Q: Which database engines are supported? A: Postgres, MySQL, MongoDB, Redis, SQL Server, SQLite, MariaDB, Elasticsearch, DynamoDB, BigQuery, Redshift, ClickHouse, DuckDB, MotherDuck, Turso, Cloudflare D1, CockroachDB, Snowflake, Cassandra, Oracle, Databricks — plus CSV, Parquet, JSON, and Excel files.

Q: Is the source code open source? A: The repository is public on GitHub. The license is not clearly listed in the README; the binary is free to download and use.

Q: Can I use my own Ollama instance? A: Yes. Tablething supports any OpenAI-compatible API endpoint, which includes self-hosted Ollama instances.

Q: How does it compare to TablePlus or DBeaver? A: Tablething adds an AI chat layer on top of standard database browsing. TablePlus and DBeaver are more mature as pure query tools; Tablething trades some depth for AI-assisted query generation aimed at non-SQL users.

Conclusion

Tablething fills the gap between a full-featured database GUI and an AI coding assistant. The ability to connect to 25+ data sources and query them via natural language — without routing data through a third-party cloud — makes it a practical option for developers who work across multiple database engines and want AI assistance without giving up local-first privacy.

The tool is actively maintained (latest release v0.15.4) and supports all three major desktop platforms. If you regularly switch between Postgres, MongoDB, BigQuery, or any of its 20+ supported backends, Tablething is worth a look.

Website: tablething.com GitHub: github.com/tablething/tablething