Google open-sources new JPEG encoder that reduces file sizes by around 35%, and is compatible with existing browsers, photo editing apps, and the JPEG standard
New algorithm is based on human psychovisual system. Images look better, too. — Google has developed and open-sourced …
Context & Ripple Effects
This encoder is the latest move in a multi-year Google campaign to shrink the web's payloads through open source: it follows the Brotli compression tool from 2015, which Google claimed beat its predecessor Zopfli by 20-26%, and lands weeks after two machine-learning image efforts — RAISR, which Google said could shrink images up to 75% and Google Brain's research on reconstructing detail in pixelated images.
What distinguishes this release is compatibility: because the output is standard JPEG readable by existing browsers and photo editing apps, the ~35% savings require no format migration — a deliberate contrast with codec transitions that stall on ecosystem adoption.
First-order effects
- Web publishers and photo-heavy sites can cut image transfer sizes by around 35% today simply by re-encoding, with no changes needed on the browser or viewer side.
- Photo editing apps and browser vendors gain the savings for free via the open-source release, since the encoder works within the existing JPEG standard rather than demanding support for a new format.
Second-order effects
- The release raises the bar for alternative image formats and compression tools: any successor must now justify itself against a baseline JPEG that is meaningfully smaller, pressuring rival codecs' adoption pitches.
- It extends Google's playbook of giving away core web-infrastructure tooling — as with Brotli — positioning the company as the de facto steward of how images travel over the network while others absorb the integration work.
Third-order effects
- If the pattern holds — Brotli, then this encoder, alongside ML-based techniques like RAISR — image delivery on the web consolidates around psychovisual and learned-compression methods, making bandwidth cost less of a constraint on rich imagery and shifting competitive advantage to whoever controls the encoding pipeline.
The trend: Google is systematically open-sourcing compression and image-processing technology — Brotli, RAISR, and now this JPEG encoder — to make smaller payloads the default state of the web.