NVIDIA® cuOpt™ is an open-source, GPU-accelerated engine for decision optimization designed to tackle large-scale problems with millions of variables and constraints.
NVIDIA® cuOpt™ is an open-source, GPU-accelerated engine for decision optimization designed to tackle large-scale problems with millions of variables and constraints.
Key features of NVIDIA cuOpt VRP Solver:
-
Dynamic Rerouting: Rerun models and adjust for changes like inoperable vehicles, traffic and weather disruptions, and the addition of new orders—all within service-level agreement (SLA) time constraints.
-
World-Record Accuracy: Experience world-record performance, achieved across Li & Lim and Gehring & Homberger accuracy benchmarks. cuOpt owns all 23 world records set in the past three years on the largest routing benchmarks verified by the Norway Institute of Science.
-
Scale Seamlessly: Scale out to 15,000 routing tasks to facilitate computationally heavy use cases.
-
Real-Time Analytics: Route 1,000 packages in 10 seconds instead of 20 minutes (120X faster)—with the same level of accuracy
-
Save time and money: Reduce travel times and fuel costs by 15 percent. cuOpt can also reduce costs by minimizing fleet size, errors, and maintenance operations. A telco company reduced cloud costs from $1.5M to $200K annually.
-
Enterprise-grade Security, Flexibility, Manageability, and API Stability: cuOpt, part of NVIDIA AI Enterprise, is a secure, production-ready AI microservice enabling flexible and seamless deployment models.
-
Reduce carbon footprint: Enterprises can reduce carbon emissions by reducing data center usage and the miles driven. One EA customer validated that they reduced the data center usage by approx. 1,367 hours (equivalent to 99% of improvement). Another food franchise customer showcased that they were able to reduce 50miles driven per store daily, saving $37M annual cost. Their reduction in miles driven equals to 50 BIL (Bipartisan Infrastructure Law) fewer grams of carbon annually.
Additional Resources
License
The container is available under Apache 2.0 License.
GOVERNING TERMS: The container and helm chart is governed by the NVIDIA Software License Agreement (found at https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-software-license-agreement/) the Product-Specific Terms for NVIDIA AI Products (found at https://www.nvidia.com/en-us/agreements/enterprise-software/product-specific-terms-for-ai-products/).
