Experimenting with the proposed Cross-Origin Storage API in Transformers.js
Get smart on it
Transformers.js is a tool that lets web developers run artificial intelligence models directly in browsers. Currently, when different websites use the same AI models or shared software resources, browsers download and store duplicate copies because cache storage is isolated by origin for security reasons. The proposed Cross-Origin Storage API would allow files to be identified and shared across different websites using cryptographic hashes instead of URLs, eliminating redundant downloads and storage. This matters because popular models like speech recognition systems are used by many different apps, and the current system wastes bandwidth and disk space by forcing repeated downloads of identical files.
Related news
Vibe coding platform Base44 launches own model as AI startups seek defensibility
Wix-owned vibe coding platform Base44 has started rolling out its own AI model — with hopes that it will eventually outperform frontier models.
DiScoFormer: One transformer for density and score, across distributions
DiScoFormer is a transformer model that estimates both the density and score of a data distribution in a single forward pass without requiring retraining for new distributions. The density describes where data points cluster, while the score, the gradient of log-density, points toward more probable regions and is used in diffusion-based generative models, Bayesian sampling, and scientific simulations. Existing methods force a trade-off: classical kernel density estimation works on any distribution but loses accuracy in high dimensions, while neural score-matching models stay accurate in high dimensions but must be retrained for each new distribution. DiScoFormer significantly outperforms kernel density estimation, cutting score error by 6.5 times and density error by more than 37 times in 100 dimensions, while generalizing to distributions with shapes and complexities not seen during training.
