Cross browser visual testing tools
Which tools actually render across Chrome, Safari, Firefox and Edge — what cross-browser coverage costs at each vendor, and why the browser matrix is usually the wrong place to spend your budget.
Cross-browser visual testing multiplies your bill by the number of engines you check. Before choosing a tool, it is worth being precise about how much that multiplication is actually buying you in 2026 — because for many teams it is the least efficient place in the budget.
What cross-browser coverage costs
| Tool | Engines | Cost of multi-browser |
|---|---|---|
| Chromatic | Chrome, Safari, Firefox, Edge | Requires Starter at $179/mo — the free tier is Chrome only |
| Applitools | Chrome, Safari, Firefox, Edge via Ultrafast Grid | Each browser is a separate checkpoint per snapshot |
| Percy | Chrome, Firefox and others | Not published — paid tiers are contact-sales |
| SmartUI | Web engines plus native iOS and Android | Free tier 2,000 screenshots; paid not published |
| Playwright | Chromium, Firefox, WebKit | Free, but a separate baseline per browser and platform |
| BackstopJS | Chrome via Puppeteer; Firefox and WebKit via Playwright | Free, self-managed |
Do you need four engines?
This is the question worth answering before the tool question, because it changes the budget by 4x.
Chrome, Edge, Opera and Brave all run Chromium. Testing Chrome and Edge separately checks essentially the same rendering engine twice — a real but narrow difference confined to form controls and a few platform behaviours. The engines that genuinely diverge are Chromium, Gecko (Firefox), and WebKit (Safari), so the meaningful matrix is three, not four or more.
Of those, WebKit is where most visual surprises still live: date inputs, scroll behaviour, backdrop filters, and flex gap handling on older iOS versions. If you cut the matrix to two engines, make them Chromium and WebKit.
- Run every pull request on one engine. Chromium, because it is what most of your users and all of your developers use.
- Run the full matrix nightly on the default branch. Cross-engine regressions are real but rarely urgent, and catching one within 24 hours is fine.
- Add WebKit to pull requests only for routes with known Safari sensitivity — anything using date pickers, sticky positioning, or backdrop filters.
That pattern typically cuts cross-browser spend by 70% while catching nearly everything. More on the pipeline shape in visual testing in CI/CD pipelines.
The tools
Best for: Large matrices where you do not want to run browser infrastructure.
- License
- Commercial
- Pricing
- Starter $667/mo billed annually · Professional and Enterprise custom
The strongest technical answer to the cross-browser problem specifically: you run your test once and the grid renders it everywhere, instead of executing the full suite per engine. Model the checkpoint arithmetic against your real route count before committing — that is where the cost lands.
- Renders one capture across many browsers and viewports in the cloud, rather than re-running your suite per browser
- Removes the need to maintain browser infrastructure in CI
- Visual AI suppresses the cross-engine antialiasing noise that makes naive matrices unusable
- Each browser counts as a separate checkpoint, so the matrix multiplies your bill directly
- Highest entry price in this comparison, on an annual commitment
- 1,000 page checkpoints on the Starter tier disappears fast across a matrix
Best for: Storybook design systems needing Safari and Firefox coverage.
- License
- Commercial
- Pricing
- Free tier is Chrome only · Starter $179/mo adds Safari, Firefox and Edge
The cleanest pricing model for cross-browser work: you buy a tier, not a multiplier. For a component library that needs to render correctly everywhere, the $179 Starter tier is often cheaper than the equivalent checkpoint arithmetic elsewhere.
- Multi-browser is a tier upgrade rather than a per-browser multiplier
- TurboSnap reduces billed snapshots by copying unchanged ones
- Unlimited parallel runs on every tier, so a wide matrix does not queue
- Cross-browser coverage starts at $179/mo — the free plan will not tell you about a Safari bug
- Component-level only; a component can pass in every browser while the page it composes is broken
Best for: Teams willing to manage baselines per engine themselves.
- License
- Apache 2.0
- Pricing
- Free
Free in licence and expensive in baseline management: the matrix that costs money elsewhere costs repository weight and process discipline here. Record baselines inside the same container CI uses, or the cross-platform mismatch will produce constant noise. Setup guide.
- Genuine Chromium, Firefox and WebKit engines, not emulation
- Projects make the matrix a few lines of config
- No per-browser billing at all
- A separate baseline per browser and platform —
name-webkit-linux.pngandname-webkit-darwin.pngare different files - That multiplication lands in your repository as committed binaries
- A developer updating a baseline locally does not update the one CI compares against
Best for: Teams whose matrix includes native mobile, not just browsers.
- License
- Commercial
- Pricing
- Free plan 2,000 screenshots · paid pricing not published
If your real cross-platform problem includes native apps rather than only browser engines, this covers ground the others do not. For browser-only matrices the alternatives are more established.
- Covers native iOS and Android through Appium, XCUITest and Espresso alongside web engines
- Works with Selenium, Cypress, Playwright, Puppeteer and Jest
- 120+ CI and collaboration integrations
- Smallest free allowance here at 2,000 screenshots, which a matrix consumes quickly
- Paid pricing is not published
- LambdaTest rebranded to TestMu AI in January 2026, so existing documentation uses the old name
Choosing
- Large matrix, enterprise budget — Applitools Ultrafast Grid, after doing the checkpoint maths.
- Storybook component library — Chromatic Starter at $179/mo, the simplest multi-browser pricing here.
- Free, and you accept baseline management — Playwright projects, recorded in a pinned container.
- Native mobile in the matrix — SmartUI.
- Not sure you need a matrix at all — you probably do not. One engine per pull request, full matrix nightly.
Frequently asked questions
What are the best cross browser visual testing tools?
Applitools Ultrafast Grid is the strongest for large matrices because it renders one capture across many browsers without re-running your suite. Chromatic has the simplest pricing, adding Safari, Firefox and Edge at the $179/month Starter tier. Playwright covers Chromium, Firefox and WebKit for free if you are willing to manage per-engine baselines.
How many browsers should visual tests cover?
Three engines matter: Chromium, Gecko (Firefox) and WebKit (Safari). Chrome, Edge, Opera and Brave all run Chromium, so testing them separately checks the same engine repeatedly. If you cut to two, choose Chromium and WebKit — Safari is where most remaining visual surprises live.
Does cross browser visual testing cost more?
Usually yes, and proportionally. Applitools counts each browser as a separate checkpoint, so a four-browser matrix multiplies your bill fourfold. Chromatic charges it as a tier upgrade instead. Playwright is free but produces a separate baseline image per browser and platform combination.
Sources
- Chromatic pricingchecked 2026-09-13
- Applitools pricingchecked 2026-09-13
- Playwright visual comparisons documentationchecked 2026-09-13
- SmartUI by TestMu AIchecked 2026-09-13