// sqlly vs tabularis

SQLly vs Tabularis

Tabularis is a free, Apache-licensed SQL workspace that went from first release to version 0.30 in nine months, with a large contributor community behind it. It keeps three drivers in the box and puts everything else in plugins, and it adds SQL notebooks and a browser mode on top. SQLly is a native Rust workbench with its drivers and admin tools built in. Both are written in Rust, but they make different bets. Here’s the honest breakdown.

Last reviewed: October 2026 (Tabularis 0.30)

The short version

SQLly

A native workbench with opinions

GPU-rendered on macOS, Windows, and Linux with no webview underneath. Thirty engines connect today through drivers built into the app, and the focus is depth: completions from real column values, layered safety guardrails, and script-first admin tools.

Tabularis

Free, hackable, community-built

A Tauri app (Rust backend, React and Monaco front end) for Windows, macOS and Linux, or a browser tab served by tabularis web. MySQL/MariaDB, PostgreSQL and SQLite are built in; 26 driver plugins, written in any language over JSON-RPC, cover the rest. SQL notebooks, a visual query builder, and an MCP server come free.

Feature by feature

SQLly’s column uses the same honest status badges as the features page and the docs. For Tabularis we list what its README, documentation and repository state as of version 0.30.

Feature comparison of SQLly and Tabularis
Capability SQLly Tabularis
Runs on macOS, Windows & Linux YesmacOS is signed and notarized; Windows and Linux are a supported beta. YesSigned macOS builds for Apple Silicon and Intel; WinGet, Homebrew, Snap, Flatpak, AUR, AppImage, deb and rpm.
How the app is built NativeRust, rendered on the GPU (Metal, Direct3D, Vulkan). How it works TauriRust backend; React and Monaco front end in the system webview.
Runs in a browser / self-hosted web app PartialThe browser build runs against your own databases through a relay you control; it is not a multi-user server. Yestabularis web serves the same UI, with Docker images, a Helm chart, password sign-in and session limits.
Free to use Free tierTwo saved servers, one cloud account, and a reminder to pay about every five hours. Pricing YesNo paid edition; sponsor-funded.
Open source NoLifetime licenses carry a source-escrow promise if SQLly is ever discontinued. YesApache 2.0.
PostgreSQL, MySQL, MariaDB, SQLite YesPlus hosted and compatible engines - 30 in all. Full list YesBuilt in. The built-in PostgreSQL driver is deprecated in favor of the PostgreSQL plugin.
SQL Server, Oracle, DuckDB, ClickHouse, libSQL/Turso YesNative drivers inside the app. Per-engine coverage PluginsEach is a separately installed driver plugin.
Redis YesCommand editor, key browser, cluster and Sentinel, streams, slow log; Valkey, KeyDB, DragonflyDB, ElastiCache and MemoryDB too. Redis PluginsTwo community plugins, one in Go and one in Rust.
MongoDB, DynamoDB, BigQuery & other non-SQL stores No PluginsMongoDB, DynamoDB, Firestore, Cassandra/ScyllaDB, Elasticsearch, BigQuery, Trino, Google Sheets and more.
Cloud sign-in (Entra ID, AWS and GCP IAM) In progressEntra ID, AWS RDS IAM and Cloud SQL IAM tokens. Azure -
SSH & other tunnels YesSSH, SOCKS, HTTP proxy, and Kubernetes port-forward. YesSSH, AWS SSM and Kubernetes tunnels.
Import connections from other tools YesDataGrip, DBeaver, pgAdmin, Azure Data Studio, VS Code, DbGate, TablePlus, Navicat and Redis Insight, with export back to most. Compatibility YesDataGrip, DBeaver, TablePlus, Sequel Ace and Beekeeper Studio.
Schema-aware autocomplete YesJoin ON clauses from real foreign keys, real column values, live GROUP BY checks. IntelliSense YesMonaco completion of tables and columns.
SQL notebooks NoDirectives in comments keep a query’s chart and layout with the script instead. SQLly directives YesSQL and Markdown cells, cross-cell variables, charts, HTML export.
Visual query builder YesTable canvas with foreign-key joins suggested. Query builder YesDrag-and-drop canvas with visual joins.
Edit table data YesA row editor generated from the table’s own metadata. Row editor YesInline and batch edits committed together.
Pivot, charts & conditional formatting in the grid YesPivot, chart, map, conditional formatting and computed columns. Pivot PartialCharts live in notebooks; the grid offers column stats.
Export results YesCSV, JSON, SQL, code, and Excel with a pivot sheet. PartialCSV and JSON from the grid; notebooks export HTML.
Data masking YesMasked in the grid, copies and exports. Masking YesPattern-based column masking in the display.
ER diagrams YesExport to SVG, Mermaid, or DBML. Relationship Graph YesInteractive, with a table picker.
Schema & data compare YesAcross connections and engines, risk-graded, with rollback scripts. Schema Compare PlannedData compare is on its public roadmap.
Execution plan viewer In progressPlain-language findings, tree, diagram and plan comparison across nine engines. Query plans YesVisual EXPLAIN graphs for PostgreSQL, MySQL/MariaDB and SQLite, with optional AI analysis.
Backup & restore YesScript-first, with the exact command shown before it runs. Backup / Restore PartialSQL dump and import.
Server activity monitoring & admin YesActivity Monitor with blocking chains, Server Dashboard, Index Analyzer, security, SQL Agent, Extended Events. Tools No
Safety guardrails YesRead-only connections, no UPDATE/DELETE without WHERE, rollback wrap, environment colors. Safety PartialPer-connection read-only mode, also applied to the MCP server.
AI assistance Rolling outOn-device, off by default, opt-in per feature. On-device AI YesText-to-SQL and explain with OpenAI, Anthropic, OpenRouter, Ollama, or any OpenAI-compatible API.
MCP server for AI agents PartialAgent queries pass the same select-only guard as everything else. YesOne-click setup for Claude Desktop, Cursor and Windsurf.
Plugins & extensions No YesDriver plugins in any language, UI extensions, theme packages, and a plugin registry.
Interface languages 2English and Spanish. 12Including Chinese, Japanese, Korean, German, French and Russian.

A dash means we couldn’t confirm it from Tabularis’s own materials - check its documentation for the current state. Tabularis ships nightly, so some “planned” rows may already have moved.

Where SQLly is different

Not “better at everything” - different in the places we care about most.

Drivers in the box

SQL Server, Oracle, DuckDB, ClickHouse, libSQL and Redis are part of the app, with live test suites against real servers, and get the same editor, plans and safety features as PostgreSQL. There’s nothing to install per engine and no driver process to keep alive.

Databases →

Native, not a webview

SQLly draws straight to the GPU with each OS’s own graphics path. Large results stream to disk, so the grid stays flat on memory whether you fetch 50 rows or 50 million.

Spill-to-disk results →

Autocomplete that knows your data

Beyond table and column names, SQLly offers a join’s ON clause from actual foreign keys, pulls a column’s real values into a WHERE clause, and flags a GROUP BY mistake as you type.

IntelliSense →

A DBA’s toolbox

Schema and data compare, activity monitor, backup and restore, index analyzer, security, SQL Agent and Extended Events live under one Tools menu. Anything destructive opens as SQL in a tab for you to read before it runs.

Tools →

A grid that does analysis

Pivot, chart, conditional formatting and computed columns over any result, set by a one-line comment that travels with the query, then exported to Excel with the pivot sheet included.

SQLly directives →

Guardrails you set once

Read-only connections that refuse file-reading and shell-escape tricks, a hard stop on UPDATE/DELETE with no WHERE, auto-wrapped transactions, and prod painted an angry red - inherited down your client, project, and environment tree.

Safety & guardrails →

When Tabularis is the better choice

Tabularis is a fast-moving, generous project with a real community behind it. Pick it if:

  • You want free and open source, with nothing held back. Tabularis is Apache 2.0 with no paid tier. SQLly’s free tier caps saved servers at two and isn’t available inside organizations with more than 100 employees.
  • Your data lives outside SQL. MongoDB, DynamoDB, Firestore, Cassandra, Elasticsearch and BigQuery all have Tabularis plugins. SQLly connects to none of them.
  • You think in notebooks. SQL and Markdown cells that pass results to each other, with charts and an HTML export, make a good runbook or analysis. SQLly has no notebooks.
  • You want a browser deployment. tabularis web runs the full UI from a container for a team. SQLly’s browser build is a relay-backed page, not a shared server.
  • You want to write your own driver. Tabularis plugins speak JSON-RPC over stdin and stdout, so any language works. SQLly has no plugin system, on purpose.
  • You’re on an Intel Mac, or want signed Windows and Linux packages today. Tabularis ships both Mac architectures and every major package manager; SQLly’s Mac build targets Apple Silicon, and its Windows and Linux builds are still a beta.
  • You’d rather work in your own language. Tabularis ships twelve interface languages; SQLly ships two.

Questions

Is Tabularis free?

Yes. Tabularis is open source under the Apache 2.0 license with no paid edition; the project is funded by sponsors. SQLly has a free tier for hobbyists and paid SQLly 1.x and Lifetime licenses for everyone else.

Which databases does Tabularis support?

Three drivers ship built in: MySQL/MariaDB, PostgreSQL and SQLite. Everything else is a driver plugin - its README lists 26, including SQL Server, Oracle, DuckDB, ClickHouse, Redis, MongoDB, BigQuery, DynamoDB, Cassandra and Trino - some maintained by the project and some by the community. SQLly builds its drivers in: 30 engines counting hosted variants, with no MongoDB or other document stores.

Is Tabularis built in Rust like SQLly?

Partly. Tabularis is a Tauri app: a Rust backend with a React and Monaco front end drawn by the operating system’s webview. SQLly is Rust all the way to the pixels, rendered on the GPU without a webview.

// try it yourself

Take the pig for a spin

Free on macOS, Windows, and Linux, no account required. Keep Tabularis open for the notebooks - the pig doesn’t get jealous.