CATCHES Launches Physics-Based AI Sizing Platform with AMIRI and Nvidia

Mar 19, 2026
CATCHES has introduced RealFit, an AI-powered sizing and virtual try-on technology for fashion e-commerce, developed with Nvidia and first implemented by AMIRI. The company also announced $10 million in funding from luxury industry investors.

In a press release, CATCHES announced the launch of its RealFit technology, a generative AI platform with physics-based garment sizing for fashion e-commerce. The debut took place at NVIDIA GTC 2026 and is powered by Nvidia’s CUDA and Omniverse platforms. The first live integration of RealFit is available on the website of fashion brand AMIRI.

RealFit allows shoppers to generate a digital twin by uploading measurements and a photo, enabling visualization of garment fit and drape with millimeter-level accuracy. The technology combines GPU-accelerated physics simulation with AI models trained on real fabric data, producing realistic representations of how clothing moves and fits.

CATCHES has raised $10 million from investors including Antoine Arnault and Natalia Vodianova Arnault, along with other industry figures. The company said it is working with additional luxury brands to roll out RealFit later in 2026.

The system uses NVIDIA’s Newton Physics Engine, RTX ray tracing, and diffusion models to simulate fabric behavior within an OpenUSD framework. This integration enables seamless data exchange across design tools such as Blender and Unreal Engine, supporting high-fidelity visualizations for online retail.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like AI Funding Brief or Daily AI Brief.

Also, consider following us on social media:

Free newsletter

AI Funding Brief

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more
Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.