Krunkit

About Krunkit

Crunch your files, not your time.

Our Mission

Krunkit is a free online image tool suite that lets you compress, convert, resize, and remove backgrounds from images — individually or in batches of up to 10 — without uploading them anywhere. Every operation runs directly in your browser using WebAssembly — the same technology that powers modern web applications.

We built Krunkit because existing image tools either require uploads to remote servers (privacy risk), charge subscriptions for basic features, or install desktop software. We believed there was a better way: bring professional-grade image codecs to the browser and let your device do the work.

What We Believe

Privacy by Design

Your image files are processed locally in your browser for the core tools. Krunkit does not upload them to our servers for compression, conversion, resize, or background removal.

Performance First

Krunkit uses established open-source codecs — mozjpeg, oxipng, libwebp, and libavif — compiled to WebAssembly so common image work can run in the browser.

Free for Everyone

No sign-ups, no subscriptions, and no watermarks. Krunkit keeps reasonable file and batch limits so browser-side processing stays reliable.

Technology

Krunkit is built with Next.js and TypeScript, and uses jSquash — an open-source collection of image codecs compiled to WebAssembly. The codecs include:

  • mozjpeg — Mozilla's optimized JPEG encoder
  • oxipng — Rust-based PNG optimizer
  • libwebp — Google's WebP codec
  • libavif — AV1 Image File Format codec

Core image processing runs on the main thread or in Web Workers within your browser tab. The selected image files are not uploaded to Krunkit servers for compression, conversion, resizing, or background removal.

Behind Krunkit

Jason Shin — Creator of Krunkit

Jason Shin — Full-Stack AI Builder

With 20 years of experience spanning IT product planning, full-stack development, and AI service architecture, Jason has led end-to-end delivery of digital products — from business modeling and strategy to hands-on engineering. His recent work focuses on practical AI services, browser-side automation, and privacy-conscious tools that solve small but frequent workflow problems.

Krunkit started from a simple frustration: every time Jason needed to compress or convert an image, the available tools either required uploading files to unknown servers — raising serious privacy concerns — or came bundled with clunky interfaces, sign-up walls, and subscription fees. None of them respected the user's time or data. So he built the tool he wished existed: one that runs the core processing in your browser and does not upload selected image files to Krunkit servers for compression, conversion, resizing, or background removal.

Krunkit is maintained as a hands-on product lab, not a generic content site. Tool copy and guides are updated from real QA findings — for example, larger compression outputs are now shown as a file-size change instead of fake savings, and background removal rejects images over 10MP before the AI model downloads.

Editorial and Testing Standards

Krunkit guides are written to support the actual tools on this site. We avoid popularity counters, fixed savings promises, and universal speed claims unless they are backed by current test data. Where results vary by image, browser, device, or format, we say so.

Before publishing tool-facing changes, we run production-style checks for linting, TypeScript, static build output, representative browser flows, mobile viewports, and privacy/trust copy. That does not make every result perfect, but it keeps the site honest about what browser-side image processing can and cannot do.

Our Approach

Most online image tools follow a simple pattern: you upload your file, a server processes it, and you download the result. This means your images travel across the internet, sit on someone else's server, and are processed by code you can't inspect. We think that's unnecessary.

WebAssembly changed what is practical in the browser. By compiling battle-tested C/C++ and Rust image codecs to run inside a browser sandbox, common image operations can happen locally without a file upload step. The result is less file exposure, no upload/download latency for the selected image, and lower operating costs than upload-based processing for many everyday workflows.

We obsess over the details that matter: accurate color handling during format conversion, intelligent transparency detection, EXIF metadata preservation where appropriate, and quality settings that balance file size with perceptual quality. Every feature is tested against real-world images — product photos, screenshots, social media graphics, portraits — to ensure consistent, reliable results.

What You Can Do With Krunkit

Compress Images

Reduce JPEG, PNG, WebP, and AVIF file sizes when smaller output is possible using professional encoders like mozjpeg and oxipng. Compare before/after quality with a visual slider, adjust quality settings, and batch process up to 10 files at once.

Convert Formats

Switch between JPEG, PNG, WebP, and AVIF with full control over quality and transparency handling. Automatic format detection identifies your source files. Transparent images get smart background color options when converting to non-transparent formats.

Resize for Any Platform

Resize to exact pixel dimensions or use built-in presets for Instagram, Facebook, YouTube, Twitter/X, LinkedIn, Pinterest, TikTok, and Discord. Lock or unlock aspect ratio, and batch resize multiple images to the same dimensions.

Remove Backgrounds

AI-powered background removal using a segmentation model that runs directly in your browser. Produces clean cutouts with soft edges for hair and fine details. Output as transparent PNG or with a custom solid background color.

Ready to crunch?

Try Krunkit now — no sign-up required.