Retool open source alternatives

The one thesis that will save you a week

The real Retool replacement question is not “which open-source tool has the prettiest drag-and-drop builder?” It is whether your team mostly needs internal CRUD apps over databases and APIs, or whether it needs scripts, workflows, approvals, and scheduled jobs that sometimes happen to need a UI. If you answer that honestly, Appsmith and Windmill stop looking like interchangeable “Retool alternatives” and start looking like two different escape routes.

Why people are leaving Retool in 2026

Retool is good software. That is part of the problem. It gets adopted by the first engineer who is sick of writing another admin panel, then slowly becomes the place where production database credentials, customer support workflows, billing operations, and “just one more internal dashboard” all go to live.

The pricing model is where the spreadsheet starts to smell. Retool’s entry paid Team tier is listed at $12 per builder/month billed monthly, with a lower-confidence USD caveat from the source data, and it also charges $7 per internal user/month billed monthly. That is two separate people-meters: one for the folks building apps, and one for the folks merely opening them.

That second meter matters. Internal tools are often valuable precisely because lots of non-engineers use them: support, ops, finance, compliance, warehouse teams, customer success, the random VP who clicks one report twice a quarter and somehow needs access forever. When every viewer becomes a line item, “we built a quick admin panel” turns into “we accidentally built a SaaS bill generator.”

Then there are workflow runs. Retool’s workflow runs are metered on top, with extra runs priced at $75 per additional 5,000/month according to the supplied pricing note. That may be fine if you are replacing a mess of cron jobs with something business-critical. It is less fine when a humble approval flow starts behaving like a coin-operated washing machine.

The other reason people look elsewhere is control. Retool often sits close to production systems, with direct credentials into databases and APIs. For some companies that is acceptable. For others, it is exactly the kind of vendor concentration that makes security reviews turn into archaeology.

The alternatives that matter

Appsmith

Appsmith is the most obvious open-source Retool alternative if your mental model is still “build internal apps fast.” It gives you a low-code environment for admin panels, dashboards, and database-backed tools, with a familiar split between UI widgets, queries, APIs, and business logic. If Retool is the thing your ops team already understands, Appsmith is the least philosophically weird replacement in this set.

Live metrics: appsmithorg/appsmith has 40798 stars, 4741 forks, 4473 open issues, license Apache-2.0, last push 2026-09-01, primary language TypeScript.

The GitHub description is refreshingly direct: “Platform to build admin panels, internal tools, and dashboards. Integrates with 25+ databases and any API.” That is the job. Not “AI-native enterprise transformation fabric,” not “agentic data plane,” just internal tools over databases and APIs. Bless.

Pick Appsmith if your Retool usage is mostly forms, tables, charts, admin screens, and workflow-ish logic glued to databases or HTTP APIs. It is the better fit when non-specialist internal-tool builders need a visual canvas and you want to self-host instead of keeping every internal panel on a vendor platform. The Apache-2.0 license is also a real advantage if your company’s legal process treats weird licensing like a haunted object.

Skip Appsmith if the thing you really need is a developer automation platform. It can build internal tools, but it is not primarily selling you a script runner, workflow engine, or backend job orchestration story. Also, 4473 open issues is not a moral failing, but it is a signal: this is a large, active project with all the noise, backlog, and “why is this edge case still open?” energy that comes with that.

Appsmith also has an official cloud option, with the supplied data listing official SaaS pricing at $15/month. That can be useful during migration, but do not confuse “has a cloud plan” with “we escaped SaaS gravity.” The point of Appsmith, for Retool refugees, is that you can run the thing yourself when the bill, data access model, or procurement process gets annoying.

Windmill

Windmill is the alternative for teams that looked at Retool and thought, “Nice buttons, but the real problem is all these scripts and workflows.” It is an open-source developer platform for turning scripts into webhooks, workflows, scheduled jobs, and UIs. Compared with Retool, it feels less like a low-code app builder that learned automation and more like an automation engine that learned how to render interfaces.

Live metrics: windmill-labs/windmill has 17753 stars, 1091 forks, 846 open issues, license NOASSERTION, last push 2026-09-01, primary language Rust.

The project’s GitHub description says: “Open-source developer platform to power your entire infra and turn scripts into webhooks, workflows and UIs. Fastest workflow engine (13x vs Airflow). Open-source alternative to Retool and Temporal.” The “13x vs Airflow” bit is exactly the kind of benchmark-flavored marketing you should verify against your own workload before tattooing it on a roadmap. Still, the positioning is clear: Windmill wants to be where your operational code runs, not just where your tables get edited.

Pick Windmill if your Retool estate is secretly a pile of operational automation: Python scripts, TypeScript jobs, scheduled syncs, incident helpers, approval flows, one-off support actions, and internal UIs wrapped around developer-owned logic. It is a strong fit for backend engineers who would rather write code than drag a widget three pixels to the left until the product manager stops sighing.

Skip Windmill if your main audience is non-technical builders who want a Retool-like canvas first and a developer platform second. Windmill can produce UIs, but its center of gravity is scripts and workflows. If the team asking for the replacement says “we need a better admin panel builder,” Appsmith is probably the cleaner conversation. If they say “we need a safer place to run privileged operational scripts,” Windmill starts looking much more interesting.

The license field here is worth being boring about: GitHub reports license NOASSERTION in the supplied data. That does not automatically mean “bad,” but it does mean you should not casually write “permissive open source” into an architecture decision record without checking the repository and your company’s legal requirements. Internal tooling already has enough footguns; licensing ambiguity does not need to be another one.

Decision framework

• If you mostly build CRUD admin panels, dashboards, and database-backed internal apps → pick Appsmith.
• If you mostly run scripts, workflows, webhooks, scheduled jobs, and operational automation → pick Windmill.
• If non-engineers need to assemble tools visually → start with Appsmith.
• If backend engineers own the logic and just need UIs around it → start with Windmill.
• If license clarity is a procurement blocker → Appsmith’s Apache-2.0 is easier to reason about from the supplied data.
• If Retool workflow metering is the pain point, not the app canvas → evaluate Windmill first.

What to watch for

Migration will not be a one-click export party. Retool apps usually contain a mix of UI layout, queries, JavaScript snippets, permissions, secrets, environment config, and deeply local assumptions about how your company operates. The annoying part is not rebuilding a table component. The annoying part is rediscovering that a support workflow depends on three SQL queries, two REST calls, a hidden transformer, and a button named “DO NOT CLICK” that everyone clicks daily.

Data connections deserve special attention. Retool often has direct credentials into production databases and APIs, and any replacement needs a deliberate plan for secrets, network access, auditability, and least privilege. Self-hosting does not magically make a risky admin panel safe; it just moves the blast radius into your own house. Congratulations, you now own the smoke alarm too.

Expect feature gaps around permissions, deployment workflows, environments, and the small conveniences your team stopped noticing. Appsmith may feel closest for classic internal apps, but you still need to test your actual components, query patterns, and auth model. Windmill may be better for automation-heavy workloads, but teams expecting a pure Retool clone may bounce off the developer-first shape.

The sane path is to pick one real Retool app and rebuild it completely, including auth, secrets, production data access, error handling, and whatever approval trail your auditors care about. Do not migrate the prettiest demo dashboard. Migrate the ugly tool that three departments depend on and one engineer is afraid to touch. That is where you find out whether you are choosing an alternative or just buying yourself a six-month rewrite with better branding.