// sqlly vs lazysql

SQLly vs lazysql

lazysql is a terminal UI for SQL databases, written in Go and modeled on lazygit: vim keys, tabs, a table tree, inline row edits, and a SQL editor, all in a binary of about 9 MB. SQLly is a desktop app with a GPU-drawn window. They solve different problems, so this is really terminal versus desktop. Here’s the honest breakdown.

Last reviewed: November 2026

The short version

SQLly

A desktop SQL workbench

One Rust codebase drawn on the GPU on macOS, Windows, and Linux. Thirty engines, IntelliSense that reads your data, a grid built for analysis, layered safety guardrails, plans, compare, and a full set of script-first admin tools. Needs a screen; it does not run in a terminal.

lazysql

A free, keyboard-only client for the terminal

An MIT-licensed Go TUI with about 4,400 GitHub stars and 50 releases since 2023. Five engines: PostgreSQL, MySQL, SQLite, SQL Server, and ClickHouse. Vim keybindings, remappable keys, TOML config with per-project overrides, and it runs anywhere a terminal does, including over SSH. Its README still calls it alpha; the latest release is v0.7.0 from October 2026.

Feature by feature

SQLly’s column uses the same honest status badges as the features page and the docs. For lazysql we list what its README, release notes, and release assets state.

Feature comparison of SQLly and lazysql
Capability SQLly lazysql
Interface DesktopOne Rust codebase, rendered on the GPU (Metal, Direct3D, Vulkan). How it works TerminalA text UI built on tview and tcell. Runs in any terminal, tmux pane, or SSH session.
Runs on a headless server over SSH NoNeeds a display. Reach remote databases through SSH tunnels or the bridge and relay instead. YesThat is what a TUI is for.
Runs on macOS, Windows & Linux YesmacOS is signed and notarized; Windows and Linux are a supported beta, on amd64 and arm64. YesRelease binaries for amd64 and arm64 on all three, plus 32-bit x86 for Linux and Windows. Editing in an external editor is Linux and macOS only.
Intel Macs NoApple Silicon only. YesA Darwin x86_64 build ships with each release.
Download size 61-76 MBmacOS dmg 65.6 MB, Windows x64 zip 60.9 MB, Linux x64 tarball 76.2 MB. 8-9 MBv0.7.0: macOS arm64 8.6 MB, Linux x86_64 9.0 MB, Windows x86_64 9.2 MB, all compressed.
Install Installerdmg, zip, or tarball from the download page, or the Mac App Store edition. PackageHomebrew, go install, GitHub release binaries, and a community AUR package.
Free to use Free tierTwo saved servers, one cloud account, and a reminder to pay about every five hours. Pricing YesNo paid tier and no account.
Open source NoLifetime licenses carry a source-escrow promise if SQLly is ever discontinued. YesMIT on GitHub.
PostgreSQL, MySQL, SQLite, SQL Server YesPlus MariaDB, Azure SQL (Entra sign-in in progress), and hosted compatibles - 30 engines in all. Full list YesSQL Server includes Azure SQL Database and Kerberos sign-in on macOS and Linux.
ClickHouse YesWith plans, dashboard, activity monitor, and compare. YesNative and HTTP protocols. Row edits run as mutations, outside a transaction.
DuckDB, Oracle, libSQL/Turso YesNative drivers. Per-engine coverage No
Redis YesKey browser, command editor, Cluster and Sentinel, slow log; Valkey, KeyDB, DragonflyDB, ElastiCache and MemoryDB too. Redis NoNoSQL support is on its roadmap.
SSH tunnels YesSSH hops, SOCKS and HTTP proxies, and Kubernetes. Engine docs PartialNo built-in SSH client. It runs commands you configure before connecting - ssh -L, kubectl port-forward - and waits for the port to open.
Where passwords live KeychainIn the OS keychain; connection settings are plain files. Config fileIn the connection URL inside config.toml, or pulled from environment variables with ${env:VAR}.
Per-project connections YesConnections are organized by client, project, and environment, and settings inherit down that tree. YesA .lazysql.toml next to your repo’s .git overrides the global config.
Import connections from other tools YesDataGrip, DBeaver, pgAdmin, Azure Data Studio, DbGate, TablePlus, Navicat, and more. Compatibility NoConnections are URLs you type or paste.
Vim-style keyboard control PartialAn optional Vim mode in the SQL editor (Toggle Vim Keybindings in the command palette); the rest of the app uses ordinary shortcuts. YesVim keys throughout, and every binding can be remapped in the config.
IntelliSense YesAlias- and scope-aware, JOIN ON from foreign keys, real column values in WHERE. IntelliSense PartialTable and column autocomplete with syntax highlighting. Long queries can go to $EDITOR.
Edit table data YesA row editor generated from the table’s own metadata. Row editor YesInsert, duplicate, edit, and delete rows; changes wait as pending until you save them.
Follow foreign keys PartialThe Relationship Graph maps the keys and IntelliSense writes JOIN ON from them, but there is no one-key jump from a cell to the referenced row. YesJump to the referenced row, or back to rows that reference it. Single-column keys only; not on MySQL.
Large results YesStreams and spills to disk past 50,000 rows or 64 MiB, flat memory at 50M rows. Streaming PartialEditor results stream and stop at 1,000 rows by default (configurable); tables page 300 rows at a time.
Pivot, charts & result diff YesPivot, charts, conditional formatting, and diffs on any result. Pivot No
Export data YesCSV, JSON, Excel, SQL, and more. Export PartialCSV files; copy rows as SQL, Markdown, JSON, or CSV.
Create tables & edit structure YesDesigners and Edit Structure. Designers NoWrite the DDL in the SQL editor. Column and index creation is on its roadmap.
Execution plan viewer In progressPlain-language findings, tree, and diagram across nine engines. Query plans No
Schema & data compare YesRisk-graded schema diffs and table diffs, with sync scripts. Schema Compare No
Session & activity monitoring YesActivity Monitor with blocking chains, top queries, Server Dashboard. Activity Monitor No
Read-only connections YesSELECT-only connections that also block file-read and shell-escape tricks. SELECT-only YesA ReadOnly flag or --read-only blocks INSERT, UPDATE, DELETE, DROP and similar statements.
Other guardrails YesNo UPDATE/DELETE without WHERE, auto-rollback, tenant checks, prod painted red. Safety -Pending row edits are previewed before they run; we couldn’t find more documented.
AI assistance & MCP YesLocal or opt-in cloud models, and a read-only MCP server for agents. On-device AI No
Themes YesClassic, Neutral, and Signature theme families. YesEight presets, including Dracula, Nord, Gruvbox, and Catppuccin, or your terminal’s own colors.

A dash means we couldn’t confirm it from lazysql’s own materials - check its repository for the current state.

Where SQLly is different

lazysql keeps you in the terminal and stays small. SQLly gives up both to put more on the screen.

A grid that does analysis

A terminal draws characters, so a TUI grid shows rows and stops there. SQLly’s grid adds pivots, charts, conditional formatting, computed columns, result diffs, and Excel export with a pivot sheet, and a one-line SQL comment saves the layout with the query.

SQLly directives →

Results with no row cap

SQLly streams results and spills to disk past 50,000 rows or 64 MiB, so a 50-million-row result uses about as much memory as a small one. You don’t have to pick a row limit before you run the query.

Spill-to-disk results →

Autocomplete that knows your data

SQLly can pull a column’s real distinct values into a WHERE clause, write a join’s ON clause from actual foreign keys, and underline a GROUP BY mistake as you type.

IntelliSense →

Guardrails past read-only

Both tools can make a connection read-only. SQLly also stops UPDATE/DELETE with no WHERE, wraps changes in transactions that roll back automatically, checks tenant scope, and paints prod red - set once on a client or environment and inherited by every connection under it.

Safety & guardrails →

DBA tools

Activity monitor, query plans, Query Store, SQL Agent, schema and data compare, index analyzer, test data, disk usage, and backup and restore live under one Tools menu. Anything destructive opens as SQL in a tab for you to read before it runs.

Tools →

More engines, Redis included

SQLly adds MariaDB, DuckDB, Oracle, libSQL/Turso, and Redis with its forks to the five engines both tools share, and documents wire-compatible services like CockroachDB, Neon, and Supabase.

Databases →

When lazysql is the better choice

lazysql is probably the better fit if:

  • You work on remote boxes over SSH. lazysql runs on the server, inside the SSH session, next to the database. SQLly can’t run without a display.
  • You live in the terminal. If your day is tmux, vim, and lazygit, lazysql fits in a pane and never makes you reach for the mouse. SQLly is a window you switch to.
  • Footprint matters. About 9 MB compressed against SQLly’s 61-76 MB, one static binary, and it installs with brew or go install. It also runs on Intel Macs and 32-bit x86, which SQLly doesn’t.
  • You want config in your repo. A .lazysql.toml beside .git, environment-variable credentials, and pre-connect commands for bastions and kubectl port-forward make it easy to script and share setups.
  • It has to be free and open source. lazysql is MIT with no paid tier. SQLly’s source is closed, and its free tier isn’t available inside organizations with more than 100 employees.
  • You mostly browse tables and fix rows. Open a table, filter with a WHERE, follow a foreign key, edit a cell, save. lazysql does that loop quickly, and you don’t need SQLly’s tools for it.

One caveat in the other direction: lazysql’s README still calls the project alpha, and it releases often, so check its repository for the current state of any feature you depend on.

Questions

Is lazysql free?

Yes. lazysql is MIT-licensed open source with no paid tier and no account. Official release builds ask once whether to send anonymous aggregate usage counts, with sending on by default; you can choose Disable, pass --no-telemetry, or set DO_NOT_TRACK=1, and builds from source or go install send nothing. SQLly is free for hobbyists, with SQLly 1.x and Lifetime licenses for everyone else.

Which databases do SQLly and lazysql support?

Both connect to PostgreSQL, MySQL, SQLite, SQL Server and ClickHouse. That is lazysql’s whole list today; MongoDB is listed as not yet supported. SQLly adds MariaDB, DuckDB, Oracle Database, libSQL/Turso and Redis, plus documented wire-compatible services such as CockroachDB, Neon, Supabase, Valkey and Amazon ElastiCache - 30 engines in all.

Can I use SQLly in a terminal or over SSH?

No. SQLly is a desktop app with a GPU-drawn window, so it needs a local display. Its sqlly-cli runs from a shell for MCP and scripting, but it is not an interactive terminal UI. To reach a database on a remote box, SQLly uses SSH tunnels or its bridge and relay from your own machine. If you need a client that runs inside an SSH session or a tmux pane, lazysql does that and SQLly does not.

// try it yourself

Point both at the same database

The quickest way to choose is to run your own queries in each. SQLly is free on macOS, Windows, and Linux, with no account required.