Loki

Loki is a visual regression testing tool for Storybook projects, built around screenshot comparisons rather than cross-browser coverage. It suits teams that want to check story changes in CI and can manage the browser or mobile simulator setup themselves.

Overview

Loki captures Storybook stories, compares them with approved reference images, and presents visual differences for review. The workflow also supports approving updated references, making it a practical way to catch unintended UI changes before they reach users.

Its scope is focused: Chrome, iOS simulator, and Android emulator captures are supported, with AWS Lambda as a remote renderer for larger suites. Loki is strongest when a team already uses Storybook and wants a free, configurable visual-checking step; it is less suitable when browser coverage or managed test infrastructure is central.

Key features

Reference images and review

Teams create baseline screenshots, run stories against them, and inspect visual diffs before approving changed references. In CI, missing reference images can fail tests, which helps make baseline setup explicit rather than silently treating new output as correct.

Capture and comparison options

Capture targets include Chrome in Docker, local Chrome, the iOS simulator, and Android emulator. Chrome on AWS Lambda is another option for large suites. Loki offers pixelmatch, gm, and looks-same diff engines; gm requires GraphicsMagick. Docker is optional and only needed for the Docker Chrome target.

Recent Storybook versions need no extra integration for non-React Native projects. Older versions may require a React configuration import, while React Native projects use loki/configure-react-native in storybook/storybook.js. Configuration can live in a loki section of package.json or in .lokirc, .lokirc.json, or loki.config.js.

CI and constraints

Loki is designed for continuous integration, but it does not start Storybook or simulator and emulator servers. Those services must already be running when tests begin, so teams need to account for that setup in their pipeline. Node 16 or later is required.

Documented animation limits include looping requestAnimationFrame animations, GIFs, SVG animations, native Lottie animations, and React Native Animated. Projects whose stories depend on these animations may need to avoid or account for them in visual checks.

Pricing

Loki is free: its plan costs nothing and is MIT-licensed open-source software. There is no free trial because the software is free. Node 16 or later is a prerequisite; the free plan has no stated seat or screenshot quota.

This makes Loki a good fit for teams able to run and maintain their own test setup, especially those who want to avoid a per-screenshot plan. The tradeoff is operational: teams must provide the running Storybook and mobile services, and the documented capture scope does not include cross-browser testing.

Platforms

Loki supports Android and iOS simulator or emulator captures, Linux, macOS, Windows, web, and self-hosted use. Browser capture targets are Chrome in Docker, local Chrome, and Chrome in AWS Lambda; cross-browser capture is not supported.

Who it's for

Loki is best for Storybook teams that want visual regression checks in CI, need mobile simulator or emulator screenshots, and are comfortable managing Node-based tooling and their own servers. It is a weaker choice for teams seeking broad browser coverage, turnkey server orchestration, or reliable visual testing of the animation types covered by its documented limitations.

Pros and cons

Pros

• Free and MIT-licensed: teams can add visual checks without a paid plan or stated screenshot cap.
• Useful mobile targets: iOS simulator and Android emulator captures extend Storybook checks beyond desktop Chrome.
• Reviewable baselines: visual diffs and explicit approval of updated references support a deliberate change-review workflow.
• Flexible setup: multiple diff engines, configuration files, Docker Chrome, and AWS Lambda provide options for different pipelines.

Cons

• Requires pipeline ownership: Loki will not start Storybook or simulator and emulator servers, so teams must arrange that lifecycle themselves.
• Chrome-only browser capture: it cannot check whether a change looks different in other browsers.
• Animation blind spots: several common web and native animation mechanisms have documented limitations.
• Node prerequisite: projects need Node 16 or later to use the tool.

Alternatives

Diffy is worth considering when a team wants a freemium service with a stated allowance of 500 screenshots per month, two environments, team management, and one month of data retention.

Happo may suit teams looking for a freemium option with 5,000 snapshots per month, Chrome, and visual and accessibility testing; its free plan has no credit-card requirement or time limit.

Percy is an alternative for teams that want a free tier with 5,000 screenshots per month, unlimited users, and unlimited projects.

SmartUI offers a free allowance of 2,000 screenshots valid for the lifetime of the plan, one month of build history, and limited user seats.

Frontend Regression Validator is another free, open-source option, licensed under Apache-2.0 and runnable locally or with Docker.

Screenshotbot may fit teams seeking a pay-as-you-go model, with a stated 10,000 free screenshots per month, up to five users, and three months of history.

Screenster is an alternative with a free plan limited to one user and one concurrent test, with monthly local or cloud run limits.

Argos is another freemium alternative; its SDKs and platform are open source and MIT-licensed.

For more options, see Visual Regression Testing Software.

Verdict

Choose Loki if your team uses Storybook and wants a free, MIT-licensed way to review visual changes in CI, including iOS and Android simulator captures. Its clearest advantage is that combination of open-source cost and focused Storybook workflow. Look elsewhere if you need cross-browser coverage, automatic server startup, or dependable capture of the documented animation types.