The Future of Image Formats - How JPEG XL and WebP2 Will Transform the Web
The Next-Gen Format Landscape - Why New Formats Are Needed
More than 30 years have passed since JPEG was standardized in 1992, and image compression technology has advanced dramatically. Multiple next-generation formats now compete as JPEG's successor on the modern web: WebP (2010), AVIF (2019), and JPEG XL (2022) are the leading contenders.
The demand for next-gen formats is driven by the explosive growth of mobile traffic. As of 2024, over 60% of web traffic comes from mobile devices, increasing the need to deliver high-quality images under bandwidth constraints. A 30-50% file size reduction compared to JPEG directly translates to faster page loads and reduced data costs.
Requirements for next-generation formats:
- Significantly better compression efficiency than JPEG (30%+ size reduction at equivalent quality)
- Support for both lossy and lossless compression
- Native alpha channel (transparency) support
- HDR (High Dynamic Range) and wide color gamut support
- Progressive decoding (gradual display)
- Fast encode and decode performance
However, format superiority isn't determined by compression ratio alone. Ecosystem maturity, browser support, toolchain availability, and backward compatibility with existing assets are the practical deciding factors.
JPEG XL Technical Features - The Universal Format Aiming to Replace JPEG
JPEG XL (ISO/IEC 18181) is a next-generation format developed by the JPEG committee, designed to completely replace existing JPEG. It was created by merging Google's PIK and Cloudinary's FUIF, officially standardized in 2022.
Key technical advantages of JPEG XL:
- Lossless JPEG transcoding: Existing JPEG files can be converted to JPEG XL without quality loss, achieving an average 20% size reduction. Furthermore, bit-perfect reconstruction back to the original JPEG is possible. This is critically important for organizations with billions of existing JPEG assets
- Progressive decoding: Displays a low-resolution preview from just the beginning of the file, with quality improving progressively as data arrives. More flexible than JPEG's progressive mode, allowing intermediate display at arbitrary resolution and quality levels
- Ultra-high resolution support: Supports up to 1,073,741,823 x 1,073,741,823 pixels. Suitable for satellite imagery, medical imaging, and other fields requiring extreme resolution
- HDR and wide color gamut: Supports both PQ (Perceptual Quantizer) and HLG (Hybrid Log-Gamma) HDR transfer functions. Color depth up to 32-bit floating point
Compression performance achieves 60%+ size reduction for lossy and 35%+ for lossless compared to JPEG. While matching or exceeding AVIF's compression efficiency, encoding speed is 5-10x faster than AVIF. Decode speed is comparable to JPEG, causing no issues for real-time display.
The reference implementation libjxl is open source, providing high-performance encoders and decoders written in C++.
JPEG XL's Browser Support Problem - Chrome's Withdrawal and Future Outlook
JPEG XL's biggest challenge is uncertain browser support. Chrome implemented JPEG XL support behind a flag in 2022, but removed experimental support in October 2023, citing "insufficient ecosystem interest." This decision sparked significant debate in the web development community.
Browser support status as of 2025:
- Safari: Full support since iOS 17 / macOS Sonoma. Standard use available across Apple's ecosystem
- Chrome: No support. No official plans for reintroduction
- Firefox: Experimental support behind a flag (
image.jxl.enabled). Timeline for default enablement is undetermined - Edge: No support due to Chromium base shared with Chrome
Chrome's non-adoption is a critical barrier to JPEG XL's widespread use. With Chrome holding approximately 65% of desktop browser share, adopting JPEG XL as a primary format for web delivery is currently impractical. However, active adoption is progressing in Apple's ecosystem (Safari, iOS apps), native applications, and image archival use cases.
When using JPEG XL for web delivery, fallbacks via the <picture> element are necessary. Combining Accept header content negotiation with CDN-level optimization to serve JPEG XL only to supported browsers is the realistic approach. CDNs like Cloudflare and Fastly are beginning to offer automatic JPEG XL conversion features.
WebP2 Development Status - Will WebP's Successor Materialize?
WebP2 was a next-generation format Google was developing as WebP's successor. It aimed to apply AV1 video codec technology to still images, targeting compression efficiency far exceeding WebP. However, development progress has slowed dramatically since late 2023, and the project's future remains uncertain.
WebP2 design goals (in-development specification):
- 30%+ compression efficiency improvement over WebP
- Advanced spatial prediction leveraging AV1's intra-frame prediction technology
- Unified lossy/lossless format
- Native animation support
- 10-bit / 12-bit color depth support
Technically, the design optimized AV1's prediction modes for still images, with flexible block size selection (4x4 to 64x64) and directional prediction (56 directions) to significantly surpass WebP's VP8-based prediction in compression efficiency.
WebP2's current status and challenges: Google's official repository (libwebp2) has seen virtually no updates since 2024, with development effectively stalled. Given Google's focus on building the AVIF (AV1-based) ecosystem, the likelihood of WebP2 reaching formal release appears low. Since AVIF already achieves many of WebP2's design goals, Google's motivation to separately release WebP2 has diminished.
From a web developer's perspective, moving forward with AVIF migration is more practical than waiting for WebP2. AVIF is supported by Chrome, Firefox, and Safari, with its ecosystem maturing rapidly.
AVIF Today - The AV1-Based Format Entering Production Use
AVIF (AV1 Image File Format) is an image format developed by AOM (Alliance for Open Media) that applies AV1 video codec intra-frame encoding to still images. As of 2024-2025, it's the most practical next-generation format, with rapidly expanding adoption for web delivery.
AVIF compression performance: Achieves 50%+ file size reduction at equivalent quality compared to JPEG. Even compared to WebP, an additional 20-30% reduction is possible. It excels particularly at low bitrates (high compression), making it powerful for thumbnails and bandwidth-constrained environments.
Browser support status (2025):
- Chrome 85+ (August 2020): Full support
- Firefox 93+ (October 2021): Full support
- Safari 16.1+ (October 2022): Full support
- Edge 85+ (August 2020): Full support
All major browsers support AVIF, with global coverage exceeding 95%. According to Can I Use data, the share of AVIF-incompatible browsers has fallen below 5%.
AVIF's weaknesses: Slow encoding speed is the primary challenge. It takes 10-50x longer than JPEG and 5-20x longer than WebP. For scenarios requiring real-time encoding (such as instant conversion of user-uploaded images), pre-encoding or asynchronous processing is essential. Additionally, maximum resolution is limited to 8192x4352 pixels (circumventable via tiling), making it unsuitable for ultra-high-resolution images. Progressive decoding is also not supported by default.
Migration Strategy - A Phased Approach to Next-Gen Format Adoption
Migration to next-generation formats should proceed gradually rather than switching everything at once. A strategy that maintains compatibility with existing image assets while serving optimal formats to supported browsers is the practical approach.
Phase 1: Building multi-format delivery infrastructure
Use HTML's <picture> element to serve different formats based on browser capabilities:
<picture> <source srcset="image.avif" type="image/avif"> <source srcset="image.webp" type="image/webp"> <img src="image.jpg" alt="description"></picture>
CDN-level automatic conversion is also effective. Services like Cloudflare Polish, imgix, and Cloudinary parse Accept headers to automatically serve optimal formats.
Phase 2: Build pipeline setup
Establish a system that automatically generates multiple formats during image asset builds. Use libraries like Sharp (Node.js), libvips, or ImageMagick to generate AVIF, WebP, and JPEG variants, with HTML templates automatically outputting <picture> elements.
Phase 3: Conditional JPEG XL adoption
Consider JPEG XL adoption for services with high Safari user share or native apps. Lossless transcoding from existing JPEG for immediate size reduction is attractive for services with large image archives. While monitoring Chrome's support status, preparing encoding pipelines for potential future full adoption is prudent.
Current recommendation: For web delivery, a three-tier configuration with AVIF as primary, WebP as fallback, and JPEG as final fallback is the safest and most effective approach.