QA Reef vs Katalon
Published seat prices, four surfaces, on-prem deployment — and a “free” tier that is not open source. That licence detail matters.
Facts about Katalon were read from katalon.com and their GitHub licence file on 30 August 2026 and are described as their published material states them — check the current version before you buy. QA Reef is not affiliated with Katalon; the name is used here only to identify the product being compared.
The short version
Katalon in one line
A broad quality platform built on Selenium and Appium, with agent, no-code and full-code authoring, published seat pricing and an on-premises deployment option.
QA Reef in one line
An open-core QA agent you deploy like a Vercel project: record or describe a flow, get a deterministic Playwright spec, run it on your hardware or ours, and get PASS, FAIL, or UNMEASURED back.
One thing worth stating carefully, because it is widely misreported: Katalon Studio is free, not open source. The licence in their own repository is a proprietary end-user agreement, not MIT or Apache. That is a legitimate business model — it is just a different one from an open core.
Side by side
| QA Reef | Katalon | |
|---|---|---|
| Open source | Open core. The qareef CLI is Apache-2.0. The automation core under packages/core is MIT; extraction to a public repo is in progress. The generated tests are plain Playwright you already own. |
No. Their katalon-studio-platform repository ships a proprietary Katalon licence agreement, not an OSI licence — free to use, not open source. The frameworks underneath (Selenium, Appium) are open source; Katalon is not. |
| Self-host | Yes. Node 22 + Playwright + a model key on your own machines, or hosted by us. Per-workspace data roots either way. | Yes. Their material describes an on-premises deployment of the Katalon Platform into your own data centre or cloud. |
| Pricing model | Not published — pre-launch. Self-hosted you pay only for your own model calls, each logged with its model id and cost. A local-model backend makes that $0. | Published, with figures. Their pricing page lists Katalon Studio at $180/seat/month monthly or $84/seat/month annual, True Automation at $200/seat/month monthly or $167 annual, and a custom Enterprise tier — plus separately priced add-ons (Runtime Engine $182/licence/month; TestCloud sessions $197/session/month). 30-day free trial. |
| Who writes the tests | You do — by recording in a hosted browser, or by giving the agent a goal. Deterministic codegen writes the Playwright spec; the model only proposes a title and assertions, stamped decided_by. |
You do, in whichever mode fits: agent mode via MCP, a no-code recorder, or full code in Groovy or Java. |
| How flakiness is handled | A heal ladder: wait-and-retry → recorded alternates → DOM heuristic → model proposal → OCR → coordinates. Every heal must clear a same-control gate. Model heals are quarantined for human review, never auto-promoted. A coordinate click, or a run that healed more than 20% of its steps, returns UNMEASURED rather than a pass. | Their docs describe two stages: classic self-healing tries other known locators for the object, and if that fails an LLM analyses page source, accessibility tree and screenshots to propose a new locator. |
| Legacy / canvas UIs | Yes. The operator reads the screen with Apple Vision OCR or a vision model, so canvas and legacy UIs with no stable locators are still driveable. | Desktop application testing (UWP, WinForms, WPF, Win32) is documented. Canvas is not. |
| CI integration | npx qareef CLI with real exit codes, a GitHub Action, and a Vercel Deployment Check that blocks a promote. |
Very broad and named: Jenkins, GitHub Actions, GitLab, Azure DevOps, CircleCI, Bamboo, TeamCity, Harness, AWS CodeBuild, Google Cloud Build, plus Jira, TestRail, Xray, qTest and the major device clouds. |
| Data ownership | Self-hosted: flows, screenshots, traces and the model-call ledger never leave your disk. Hosted: per-workspace data root, API tokens hashed at rest. | Hosted, or on-premises in your own infrastructure. |
Choose Katalon if…
- You need web, mobile, API and Windows desktop from one vendor.
- You want a published per-seat price and a 30-day trial today.
- You want on-prem deployment of a mature platform, not a pre-launch one.
- Your CI or ALM stack is unusual — their integration list is the longest here.
Choose QA Reef if…
- You want an actually open core, not free-as-in-beer under a EULA.
- You want Playwright rather than Selenium underneath.
- You want model-proposed heals held for review instead of applied.
- You need OCR operation for canvas and map UIs.
Where QA Reef falls short
Katalon covers four surfaces, ships on-prem, publishes prices and integrates with everything. We cover one surface and have no price. The only axis where we lead is licensing and provenance.
Sitewide caveats, on every page: QA Reef is pre-launch. No customers, no case studies, no published price, no SOC 2 report. Mobile (native app) testing is not supported. There is no managed human-QA service.
Questions
Is Katalon open source?
No. Katalon Studio is free to use, but the licence file in their own public repository is a proprietary Katalon end-user agreement, not MIT or Apache-2.0. The Selenium and Appium frameworks it builds on are open source; Katalon itself is not.
What does Katalon cost?
Their pricing page publishes $180/seat/month monthly or $84/seat/month annual for Katalon Studio, and $200 / $167 for True Automation, with add-ons priced separately and Enterprise via sales.
Selenium or Playwright — does it matter?
For most flows, less than people claim. It matters for auto-waiting, tracing and modern-app ergonomics, which is why QA Reef generates Playwright. Katalon lists Playwright among its integrations too.
Which self-healing is safer?
Katalon’s escalates to an LLM and applies a locator. QA Reef’s stops short: a model heal is used for one run and parked for a human. Safer, and more work.
Sources
Read 30 August 2026. Every statement about Katalon above comes from one of these; nothing is inferred, estimated, or taken from a third-party comparison.
Comparing more than one?
Every rival we have written up, in one table — QA Wolf, the codeless platforms, the monitoring tools, the agencies, and rolling your own Playwright.
See the full comparison index →