PNG Compressor — Up to 80% Smaller
Compress PNG files locally on Mac or Windows — no upload, no file size cap, no quality loss. Drop a folder, get back optimized PNGs in seconds.
Quick answer
TinyPixels compresses PNG files using lossless re-encoding (oxipng-based) by default, and an optional lossy palette-reduction mode for even smaller files. Both run 100% on your device — no upload, no account, no size limit.
How to compress a PNG file
Download and open TinyPixels
Free to install on Mac or Windows — no account needed to start compressing.
Drop your PNG file or folder in
Drag a single image or an entire folder of PNGs into the app window.
Choose lossless or lossy mode
Keep pixels identical, or enable palette-based quantization for extra savings.
Compress and collect the output
Optimized files land in a separate output folder — your originals are never overwritten.
How PNG compression works
PNG is a lossless format by design, but the default encoders used by most design and screenshot tools leave a lot of size on the table — unused metadata, suboptimal filter selection, and uncompressed color palettes. A dedicated PNG compressor re-encodes the same pixel data more efficiently without touching image quality.
For files that can tolerate a small quality tradeoff, lossy PNG compression reduces the color palette (quantization) before re-encoding, often cutting size by another 30–50% on top of lossless gains. TinyPixels gives you both modes with a quality slider, so you choose the balance per project.
Lossless by default
Every pixel stays identical. Only redundant metadata and encoding overhead gets stripped.
Optional lossy mode
Palette-based quantization for maximum savings, with a quality slider for full control.
Batch & folder watch
Drop hundreds of PNGs at once, or set a folder to auto-compress new files.
No file size limit
Compress a 5KB icon or a 500MB screenshot — no caps, no throttling.
TinyPixels vs online PNG compressors
| Feature | TinyPixels | Online tools |
|---|---|---|
| Works offline (no upload) | ✅ Always | ❌ Never |
| Max file size | ✅ Unlimited | ❌ Usually 5–75MB |
| Batch size | ✅ Unlimited | ❌ 20 files or fewer |
| Lossy + lossless modes | ✅ Both | ⚠️ Varies by tool |
| Folder auto-compress | ✅ Yes | ❌ No |
| Privacy | ✅ Files never leave your machine | ⚠️ Uploaded to a server |
Who gets the most out of a dedicated PNG compressor
Developers shipping app or web assets
Icons, UI graphics, and screenshots embedded in a codebase directly affect bundle size and repo size — pre-compressing before commit compounds savings across every clone and deploy.
Designers exporting from Figma or Sketch
Design tool exports are frequently far larger than needed, carrying unused metadata and suboptimal encoding by default.
Documentation and content teams
Technical docs full of annotated screenshots accumulate size fast — batch compression keeps a growing documentation site's image weight under control.
Anyone with a large existing PNG library
A one-time bulk pass over an existing folder of images, photos, or assets can reclaim meaningful disk space with zero quality tradeoff in lossless mode.
Getting the best results
A photo saved as PNG won't compress like a screenshot
If a photographic image is compressing far less than expected, that's the format-content mismatch showing up — consider converting to JPEG or WebP instead, which are built for photographic content.
Start with lossless, add lossy only if you need more
Run lossless compression first — it's free savings with zero visual risk. Only reach for lossy quantization if you need the additional reduction and can tolerate a palette-based quality tradeoff.
Batch a representative sample before committing to a setting
When testing a new lossy quality level, compress a handful of typical images first and inspect the output before running it across a full library.
Re-compressing an already-compressed PNG yields little
If a file has already been through lossless optimization, running it through again won't meaningfully shrink it further — that's expected, not a bug.
Frequently asked questions
What is the best PNG compressor?
TinyPixels is a native desktop PNG compressor for Mac and Windows. It combines lossless optimization (removing unnecessary metadata and re-encoding chunks) with optional lossy palette reduction, typically cutting PNG file sizes by 60–80% with no visible quality loss.
Does compressing a PNG reduce quality?
Not necessarily. Lossless PNG compression only strips redundant data and re-encodes the image more efficiently — pixels stay identical. TinyPixels also offers an optional lossy mode (palette-based quantization) for even smaller files, with a quality slider so you control the tradeoff.
How much smaller can a PNG file get?
Typical results with TinyPixels range from 60–80% smaller, depending on the source image. Screenshots and UI mockups with large flat-color areas compress the most; photographic PNGs compress less than JPEG/WebP would for the same content.
Is there a PNG compressor that works offline?
Yes. TinyPixels runs entirely on your machine with no internet connection required. Unlike browser-based PNG compressors, your files never leave your disk.
Why do some PNGs compress much more than others?
PNG compression efficiency depends heavily on content. Screenshots, UI mockups, and icons with large flat-color areas and few unique colors compress dramatically, since there's a lot of redundancy for the algorithm to exploit. Photographic PNGs, or images with fine noise and gradients (like screenshots of photos or AI-generated images), have less redundancy and compress far less — this is expected behavior, not a sign the tool isn't working.
Should I use lossy or lossless PNG compression?
Use lossless for anything where pixel-perfect accuracy matters — logos, icons, screenshots used for documentation, technical diagrams. Use lossy (palette-based quantization) when a small quality tradeoff is acceptable in exchange for meaningfully smaller files, which is common for general web images and app assets where nobody's zooming in to inspect individual pixels.
Does PNG compression remove transparency (alpha channel)?
No, both lossless and lossy PNG compression in TinyPixels preserve the alpha channel. Lossy quantization reduces the color palette but keeps per-pixel transparency intact, so images with transparent backgrounds compress the same way as opaque ones.
Compress your PNGs locally, right now
Free to start. No credit card, no account, no cloud. See Pro pricing →
Related tools
JPEG Compressor
Compress JPEG photos locally with quality control.
Batch Image Compressor
Compress thousands of files in one pass.
TinyPNG Alternative
Local compression — no uploads, no cloud.
PNG vs WebP
When to switch formats entirely instead of just compressing.
Image Compressor Hub
Every format, platform, and framework in one place.
Offline Image Compressor
For anyone prioritizing privacy above all else.
WebP Converter
Convert any format to WebP locally.
PNG to WebP
Convert PNG to WebP locally, keeping transparency.
Read more: How to reduce image file size without losing quality → · Why screenshots and AI images are so huge →