Visual Regression Testing Online
A zero-setup screenshot diff for manual visual regression checks — paste two builds, see every changed pixel and a diff percentage.
100% free · No sign-up · Runs in your browser
Diff two screenshots now →Automated visual regression suites like Percy, Chromatic or BackstopJS are great once they're wired into CI — but plenty of visual checks don't justify that setup: verifying a hotfix didn't shift the layout, triaging a "something looks off" report, or comparing a staging screenshot against production. For those, you just need two screenshots and a pixel diff, right now.
That's what this tool does. Paste the "before" and "after" screenshots straight from your clipboard, hit Pixel Diff, and changed pixels light up in red over a grayscale base. A threshold slider filters font anti-aliasing and rendering noise, cluster boxes outline each changed region, and a diff percentage gives you a single number to judge the change by — 0.00% means pixel-identical.
If the two captures are slightly offset — a scrollbar, an extra toolbar — use Overlay Align to nudge one over the other with 1px arrow-key precision before diffing. When you're done, export the diff view as a PNG and attach it to the bug report or pull request as evidence.
How to use
- 1Capture your "before" and "after" screenshots — from two builds, two environments, or your CI artifacts.
- 2Paste each one into a zone with Ctrl+V (no saving to disk needed) and hit "Pixel diff".
- 3Read the diff percentage, raise the threshold to mute rendering noise, enable cluster boxes to locate each change, and export the diff PNG.
Why use this tool
Diff percentage readout
One number to reason about: the share of pixels that changed, live-updated as you tune the threshold.
Threshold noise control
Filter sub-threshold deltas like font smoothing and JPEG noise so only real regressions stay highlighted.
Cluster boxes
Each changed region gets a numbered box — jump straight to what moved instead of scanning the whole page.
Zero setup, fully local
No pipeline config, no baseline management, no account. Screenshots are processed in your browser and never uploaded.
Honest comparison: this tool vs CI visual regression suites
| View Multiple Images (this tool) | Percy / Chromatic / BackstopJS | |
|---|---|---|
| Setup required | None — open the page | CI config, baseline management, per-project setup |
| Runs automatically on every commit | No — manual checks | Yes — their core strength |
| Time to first diff | Under a minute | Hours of initial setup |
| Diff controls | Threshold, intensity, cluster boxes, diff % | Varies; typically threshold + ignore regions |
| Where screenshots go | Stay local in your browser | Uploaded to their cloud (Percy/Chromatic) |
| Price | Free, no account | Free tiers; paid per-screenshot at scale |
Use this tool for spot checks, triage and one-off comparisons — when the question is "did these two builds render the same?" and you want the answer in seconds without touching your CI.
Use Percy, Chromatic or BackstopJS when you need visual regressions caught automatically on every commit across a whole test suite. This tool doesn't automate anything — by design.
Frequently Asked Questions
Can this replace Percy or BackstopJS?
No — it has no CI integration, baselines or automation. It replaces the manual side: quick spot checks, triage, and verifying individual fixes without pipeline setup.
How do I get a pass/fail signal?
Use the diff percentage: set a threshold that mutes rendering noise, then treat 0.00% as pass. Cluster boxes show exactly where any remaining difference is.
The two screenshots are offset by a few pixels — can I still compare?
Yes. Overlay Align mode lets you shift one image with arrow keys (1px, or 10px with Shift) until they line up, and the diff then runs on the aligned pair.
Are my UI screenshots uploaded anywhere?
No. The diff runs entirely in your browser — nothing is sent to a server, which also makes it safe for unreleased UI.