NVIDIA details retail AI projects spanning shopping assistants and warehouse simulation
NVIDIA highlighted retail deployments involving Shopify and Otto Group, alongside three early-access tools for store operations and model training.

NVIDIA outlined retail AI projects at GTC Berlin spanning merchant assistants, product search and warehouse planning. The examples include specialized open models and simulations of physical operations.
Shopify adapted NVIDIA Nemotron 3.5 Lightning for a Sidekick component that retrieves store information. NVIDIA said Shopify's internal testing showed slightly better accuracy than a leading frontier model, with serving costs reduced by 90% and responses arriving seven times faster.
Otto Group expanded its warehouse automation initiative to its largest fulfillment facility in Haldensleben, Germany. Built with Reply using NVIDIA Omniverse and Google Cloud, the system lets logistics teams evaluate operational changes virtually before implementing them physically.
Nebius is demonstrating three NVIDIA retail blueprints available through early access. They cover store management assistance, inventory planning through simulation and a training process that uses operational results to improve models. These provide additional starting points for retailers developing their own systems.
The day's biggest AI news in one short email, every morning. Plus an instant heads-up when a reset or a big release lands.
More on NVIDIA
- Oct 22NVIDIA expands Cosmos tools for world model training and simulationOfficial
- Oct 22Nvidia B300 servers reportedly linked to China Mobile computing hubRumor
- Oct 22NVIDIA launches Minerva to detect problems across AI data centersOfficial
- Oct 21Nvidia details Kumo Relational model and claims gains on database prediction testsOfficial
- Oct 21Nvidia’s financing role grows as AI infrastructure companies face funding pressureNews
- Oct 21Recursive releases Flint coding model for local hardwareOfficial