Free tools Windows power users keep installed
One-click scans. No signup required.
Applitools reported completing 100 visual-test runs in 21 seconds in a Storybook demonstration using Eyes and Ultrafast Grid. The batch exercised one Storybook story across 50 viewport sizes and two browsers—Firefox and Google Chrome. It was a vendor-reported result for that particular workflow, not proof that 100 arbitrary end-to-end tests can finish in 20 seconds.
What the 20-second result means
The title rounds Applitools’ reported 21-second result. In its 2019 demonstration, the company configured 100 visual runs for a single Storybook story across 50 unique viewports and two browsers. The reported time covered the SDK and server workflow, including snapshot upload, rendering, baseline creation, and result reporting.
That is broad visual coverage of one component or story, not 100 separate user journeys through a product. A test that signs in, navigates through an application, submits forms, and checks business outcomes does different work. Parallelizing rendering does not make those interactions disappear.
The runtime comes from Applitools’ own demonstration; the cited account does not provide an independent reproduction or a controlled comparison with other test frameworks. Treat 21 seconds as an example of what that configured workflow reported, not as a general benchmark or a runtime guarantee.
#1 Best Overall
How Storybook visual testing reduces the work
Collect snapshots locally
In the described workflow, the Eyes Storybook SDK reads the stories and runs them through the Storybook engine. It collects DOM snapshots locally and uploads them to Ultrafast Grid. This initial work and the upload are part of the overall process; the cloud rendering stage is not the entire test run.
Render configurations in parallel
Ultrafast Grid uses the requested browser and viewport configurations to render the uploaded snapshots, capture images, and send them for visual analysis. The speed comes from distributing those renders across configurations rather than sequentially opening a full local browser session for every combination.
Review visual results
Visual testing checks rendered output against baselines and reports differences for review. It answers whether a story’s appearance changed under the selected configurations; it does not by itself establish that an application workflow behaves correctly or that a change is functionally acceptable.
What is and is not being tested
| Question | What the demo covers | What it does not establish |
|---|---|---|
| How many runs? | 100 visual runs of a single Storybook story. | 100 independent product workflows or 100 unrelated pages. |
| Which environments? | 50 unique viewport sizes across Firefox and Google Chrome. | Every browser, device, operating system, or viewport a real product might support. |
| What work happens? | Local story processing and snapshot upload, followed by cloud rendering, screenshot capture, and visual analysis. | That the cloud service replays each user interaction in a complete end-to-end test. |
| How strong is the timing evidence? | Applitools reported 21 seconds for its demonstration. | An independently verified runtime or a promise for another project’s workload. |
How to apply the idea to your own test suite
First decide whether the test is visual or behavioral. Storybook stories and other stable, isolated UI states are good candidates for snapshot-based visual coverage. Tests that verify authentication, navigation, API effects, or user flows still need interaction and functional checks; a visual snapshot service is not a substitute for those assertions.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problems- Choose representative states. Identify stories or page states that matter, including meaningful variations such as empty, populated, error, and responsive layouts.
- Select browser and viewport coverage. Specify the combinations your users and support policy require. More combinations increase coverage, but the demo’s 50 viewports and two browsers are not a universal recommendation.
- Establish and review baselines. Ensure the reference rendering represents the intended appearance, then review detected differences rather than automatically treating every change as a defect.
- Keep functional checks alongside visual checks. Use interaction tests for behavior and visual comparisons for appearance. Decide which checks need to block a change and who reviews baseline updates.
- Measure your own end-to-end time. Record local story processing, upload, remote rendering, and result reporting separately. Compare equivalent runs and configurations before making a performance claim about your suite.
Applitools published a separate example reporting four seconds for rendering after upload in a smaller two-browser, two-viewport case. That is a different vendor-reported scenario; it should not be combined with or substituted for the 21-second 100-run result.
Using a screenshot API for a different task
A screenshot API can capture a page on request, but that is not the same thing as running Storybook visual tests: it does not replace story selection, test assertions, baseline management, or cross-configuration comparison. For standalone page screenshots, ScreenshotNeo is an alternative to try first: it removes supported consent banners, newsletter popups, and chat widgets before capture, bills only clean shots, and has a free tier of 1,000 shots per month with no card.
Or skip the browser setup
For a one-off page capture, make one request to ScreenshotNeo. This captures a screenshot; it does not run the Storybook visual-test workflow described above. See the ScreenshotNeo API documentation for request options and response details.
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webpSupported consent banners, popups, and chat widgets are removed before the shot. Bot checks, blank pages, and failed loads are not billed. An MCP server provides screenshot tools for AI agents, and the free plan includes 1,000 screenshots per month without a card; paid plans start at $5 for 3,000. Sign up for 1,000 free screenshots a month, with no card required.
Common misreadings and practical checks
- “100 tests” means 100 workflows. Not in this demonstration: it used a single Storybook story across many visual configurations.
- “20 seconds” is a guaranteed suite time. The source reports 21 seconds for its specified batch. Your story complexity, upload, configuration set, and analysis path can differ.
- “Parallel” means no local work. The SDK still processes stories and uploads snapshots before remote rendering and analysis.
- More viewports always mean better testing. Select configurations that match actual support and risk; the cited demo does not establish an optimal set.
- Visual comparisons replace end-to-end tests. They answer different questions. Retain interaction and functional assertions wherever behavior matters.
Frequently Asked Questions
Can any test framework run 100 UI tests in 20 seconds with parallelism?
No general runtime follows from this demonstration. Its figure applies to the specific Storybook visual-testing workflow and configuration described above.
Does the reported 21 seconds include rendering and results?
Applitools describes the reported batch time as covering the SDK/server workflow, including upload, rendering, baseline creation, and result reporting.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
