Business · Updated 22 Oct, 10:30 am IST
NVIDIA and Shopify use open models to cut retail inference costs and speed responses
Why it matters for readers: Explains how AI model tuning and system-level optimizations can make shopping assistants and recommendations faster and cheaper.
- Shopify post-trained an open model (Nemotron 3.5 Lightning) using NVIDIA Megatron Bridge and NeMo RL for its GraphQL subagent in Sidekick.1
- On Shopify’s internal benchmark the post-trained model surpassed a leading frontier model in accuracy while cutting serving costs by 90% and delivering responses 7x faster.1
- NVIDIA CCluster vLLM inference optimization delivered up to 4.5x faster performance and up to 12x lower inference cost across five additional Shopify use cases.1
- European retailers are applying NVIDIA-accelerated open and frontier models across agentic commerce, digital product discovery and retail R&D, and simulating multi-robot fleet collaboration in digital twins.1
Get a brief like this every morning
Uzha reads hundreds of sources and gives you the stories that matter for your work, with every source linked. Free.
Get started