QOI Format Guide

Available Conversions

Convert to QOI

About QOI Format

QOI (Quite OK Image Format) is a lossless image format created by Dominic Szablewski in 2021, designed to provide fast encoding and decoding speeds while achieving compression ratios comparable to PNG. The format's defining characteristic is its extreme simplicity: the entire specification fits in approximately 300 lines of C code, making it one of the simplest image compression formats ever designed. QOI supports 24-bit RGB and 32-bit RGBA pixel data, handling both opaque and transparent images with ease. Despite its simplicity, QOI achieves compression ratios that are competitive with PNG for most types of images, while encoding 20-50 times faster and decoding 3-4 times faster. The format uses a combination of run-length encoding, difference encoding, and a small hash table of previously seen pixels to compress image data efficiently without any lossy transformation. QOI files begin with a 14-byte header containing the magic bytes "qoif", image dimensions, channel count, and color space information, followed by the compressed pixel data and an 8-byte end marker.

History of QOI

QOI was created by Dominic Szablewski, a German game developer and programmer, who published the format specification and reference implementation on November 24, 2021. Szablewski's goal was to design an image format that was "quite OK" — not necessarily the best at any single metric, but good enough at everything while being trivially simple to implement. The format went viral almost immediately upon release, garnering widespread attention on Hacker News, Reddit, and social media, with developers praising its elegant simplicity and surprisingly competitive performance. Within weeks of its release, QOI implementations appeared in dozens of programming languages, including C, C++, Rust, Go, Python, JavaScript, Java, C#, Zig, and many others. The format's simplicity meant that a complete encoder and decoder could be written from scratch in a single afternoon, leading to rapid adoption in hobbyist and indie game development communities. Pillow, Python's primary image processing library, added QOI support in version 10.0 (released July 2023), bringing QOI to the broader Python ecosystem and enabling easy integration with existing image processing workflows. By 2024, QOI had established itself as a viable alternative to PNG for applications where encoding and decoding speed are critical, particularly in game engines, real-time rendering pipelines, and automated image processing systems.

Key Features and Uses

QOI uses four simple operations to compress pixel data: QOI_OP_RUN encodes runs of identical pixels (up to 62 consecutive repetitions in a single byte), QOI_OP_INDEX references one of 64 previously seen pixels using a hash-based lookup table, QOI_OP_DIFF encodes small differences from the previous pixel (2-bit deltas per channel), and QOI_OP_LUMA handles larger differences with 6-bit green delta and 4-bit red/blue deltas relative to green. When none of these compact representations apply, pixels are stored with QOI_OP_RGB (4 bytes) or QOI_OP_RGBA (5 bytes) for full pixel data. This combination of simple operations makes the codec extremely fast: there are no complex transforms, no entropy coding, no block processing, and no memory-intensive operations. The format supports sRGB and linear color spaces, specified in the file header. QOI files are typically 10-30% larger than optimized PNG files for photographic content but can be comparable or even smaller for images with large flat-color areas, gradients, or repeating patterns common in game graphics and user interface elements. The format's constant-time per-pixel processing makes encoding and decoding speed highly predictable, with no worst-case slowdowns.

Common Applications

QOI is primarily used in game development, where its fast encoding speed enables real-time texture compression and screenshot capture without frame drops or stutter. Indie game developers and game engine authors have adopted QOI as an alternative to PNG for storing game assets, level data, and sprite sheets, benefiting from dramatically faster load times during development iteration. Automated image processing pipelines use QOI as an intermediate format when images need to be written and read frequently between processing stages, as the format's speed advantage over PNG translates to significant time savings in batch operations. Real-time applications such as screen recording, video frame extraction, and live image preview systems use QOI to compress frames with minimal latency. The format has found adoption in embedded systems and IoT devices where computational resources are limited, as QOI's minimal memory footprint and simple implementation make it practical for resource-constrained environments. Screenshot tools and image viewers use QOI for fast temporary storage, and scientific computing applications leverage QOI for rapid capture and storage of visualization output. Converting QOI to widely supported formats like PNG, JPG, or WebP is essential for web publishing, client delivery, and sharing with users whose software may not yet support QOI natively.

Advantages and Disadvantages