What Can a GPU-Accelerated Solver Do for You?
Gurobi applauds the open-sourcing of NVIDIA’s cuOpt PDLP to advance first order methods for large-scale optimization.
We’ve been hard at work researching the newest technology, with exciting findings that reveal significant speed-ups on some large-scale instances. With our GPU-based PDHG implementation now available for testing, we invite you to submit your models and explore the potential performance improvements.
New Options for Solving Giant LPs
GPUs have now made first-order methods a compelling option for solving massive linear programming problems—but when are they the right choice? Join Dr. Ed Rothberg as he explores alternative approaches, key tradeoffs, and how these methods can drive real advancements in optimization.
NVIDIA CEO Jensen Huang GTC Keynote
Watch NVIDIA CEO Jensen Huang unveil the next wave of AI and accelerated computing at GTC 2025, including the open-source release of cuOpt’s PDLP, which will help bring more GPU-friendly algorithms to linear programming.
Using GPUs to Solve LPs: What’s In It For Me?
Explore our latest benchmark results for new GPU-enabled algorithms for barrier and PDHG.
Explore our latest benchmark results for new GPU-enabled algorithms for barrier and PDHG.
With Gurobi, you can unlock faster LP performance using a GPU-accelerated solver.
Gurobi’s GPU-based PDHG implementation can significantly speed up solve times for some large-scale instances.
Contact us today to submit your models for testing.
Join Industry Leaders Who Rely on Gurobi for Smarter, Faster, Decisions
