Safari shipped support for JPEG XL in 2023, and Chrome still refuses to enable it by default. That single split tells you almost everything about the strangest format war in web development right now. On paper, JPEG XL looks like the obvious successor to the format that has dominated the web since 1992. In practice, AVIF is winning the deployments while JPEG XL wins the benchmarks. Understanding why requires looking past spec sheets and into how browsers, CDNs, and content pipelines actually work.
Start with what each format is good at. AVIF descends from the AV1 video codec, which means it inherits a mature, hardware-accelerated compression engine that’s already deployed across billions of devices. It consistently beats WebP and legacy JPEG on file size at equivalent quality, often by substantial margins, especially at lower bitrates. That efficiency is why Netflix, Google, and a growing list of CDNs push AVIF aggressively. If your goal is shaving bytes off a photo-heavy page, AVIF delivers.
JPEG XL takes a different philosophical approach. It was designed with backward compatibility in mind, allowing a single file to contain both a traditional JPEG fallback and the enhanced version. It also handles lossless recompression of existing JPEGs, which means you can convert an entire image library without visible quality loss and cut file sizes meaningfully. For archival and photography workflows, that property is close to magic. It also decodes progressively and handles high bit depths and wide color gamuts cleanly, which matters for HDR content.
So why hasn’t JPEG XL taken over? Browser politics, mostly. Google’s Chrome team has repeatedly cited limited ecosystem interest and questions about whether the incremental gains justify the maintenance burden. Critics point out that Google has a competing interest in AVIF and WebP, both of which it helped develop, and that the decision looks less like a neutral engineering call and more like a strategic one. Mozilla has flip-flopped, and while Firefox now supports JPEG XL behind a flag, it isn’t on by default everywhere. Safari is the lone major holdout that ships it broadly.
That leaves developers in an awkward spot. The pragmatic answer for most production sites is AVIF with a WebP or JPEG fallback, served through content negotiation. It’s well supported, tooling is mature, and the compression wins are real. JPEG XL remains the better format for specific jobs: recompressing existing JPEG libraries without generational loss, handling wide-gamut and HDR images efficiently, and any workflow where a single file needs to serve both old and new clients.
There’s also a performance nuance people often miss. AVIF’s encoding is slow and computationally heavy. Generating AVIF at scale can strain build pipelines, and while hardware decode is increasingly common, encode remains largely software-based. JPEG XL tends to encode faster and decode more gracefully on constrained devices, though real-world benchmarks vary widely depending on the encoder and settings. Neither format is uniformly faster; it depends on what you’re optimizing for.
The security and patent picture adds another wrinkle. AVIF rides on AV1, whose licensing has been contentious, though the format itself is royalty-free in intent. JPEG XL was designed to avoid the submarine patent problems that helped sink JPEG 2000 and, to a lesser extent, complicated HEIC adoption. For companies wary of licensing exposure, that’s a meaningful consideration that rarely shows up in glossy comparison charts.
None of this is settled. Browser support can shift with a single release, and the format that looks doomed today can become the default tomorrow. What’s clear is that the web doesn’t need one universal winner. AVIF is the safer bet for general deployment right now, and JPEG XL is the more elegant technical solution that the ecosystem hasn’t fully embraced. Developers who bet entirely on either one are taking a risk. The smarter move is building image pipelines that can adapt, serving whatever the client actually supports rather than committing to a format because a benchmark chart looked good.
The format war will likely resolve the way most do: not by technical merit alone, but by whoever controls the browsers. And right now, that’s a fight JPEG XL is losing for reasons that have surprisingly little to do with how good it is.