Lyric Leverages NVIDIA cuOpt to Elevate Supply Chain AI

AI Becomes a Supply Chain Workhorse

For major consumer packaged goods (CPG) companies, distribution is one of the highest value yet most computationally challenging problems to solve. Lyric is integrating its composable supply chain decision intelligence capabilities with NVIDIA cuOpt, a GPU-accelerated solver suite. This brings AI to the supply chain optimization landscape, delivering results that defy conventional limits at lightning-fast speeds.

cuOpt’s power of advanced compute doesn't just improve existing processes—it fundamentally changes what's possible. It empowers supply chain modeling and planning teams to test large-scale network simulations previously impossible due to computational challenges, opening up multiple avenues to continually optimize cost, efficiency and service.

As Ganesh Ramakrishna, Founder and CEO of Lyric explains:

"Our early results have been nothing short of phenomenal. We're talking about supply chain planners testing a hundred times more scenarios in the time it used to take for a single run and achieving improvements of 150–200 basis points at 10-20x speed, directly attributable to the NVIDIA stack.

This is a genuine game-changer. Lyric is excited to work with NVIDIA in bringing this breakthrough to the supply chain ecosystem. In Lyric's view, this is a monumental technological contribution because the supply chain is what keeps the world in motion."

Intelligence Platform Meets Computational Power

Lyric Studio's comprehensive four-layer platform provides the enterprise foundation—connecting data sources through its Data Layer, deploying specialized algorithms via its Algorithm Layer, enabling no-code solution building with its Workflow Layer, and delivering insights through configurable interfaces in its Experience Layer. When this rich architecture is supercharged by NVIDIA cuOpt technology, the result is transformative: businesses gain both the analytical agility and computational speed required to solve today's most challenging supply chain problems in a way that fits and adapts to business needs like never before.

Compute to Compete: The Breakthrough That Changes Everything

Traditional CPU-based optimization tools have struggled to scale in supply chain decision-making, often taking hours to solve complex routing and scheduling problems. This limitation narrows the scope of the analysis to just one or a few scenarios, restricting the optimization potential of running multiple what-if scenarios. With ever growing uncertainties, planners and decision makers want to have multiple choices and not just one optimal answer. This shift has driven the demand for always-on, dynamic supply chain decision-making. The integration of Lyric with NVIDIA cuOpt unlocks new possibilities.

Our integration of NVIDIA cuOpt is transformative:

• 240x Faster Routing: Solutions that once took 4 hours now reach high-quality results in just 1 minute, with ongoing optimization delivering even greater improvements.

• Enhanced Solution Quality: Consistently achieving a 200-basis point improvement over CPU-based alternatives, which is huge in terms of dollar value

• Scenario Explosion: The ability to evaluate 100+ scenarios in the time previously required for a single run

• From Batch to Interactive: Moving from overnight batch processing to interactive, near real-time decision support

For supply chain leaders, this shifts GPU-accelerated optimization from a technical curiosity to a must-have competitive capability.

From Theoretical to Practical: Real-World Applications

The value of this computational breakthrough becomes concrete when tailored to specific supply chain challenges across:

Seed-to-Table Optimization: Supply chains no longer operate in the world of silos. Leading consumer products supply chains are building end-to-end optimization models that include commodity/material inputs to customer delivery requirements. Such huge models were previously impossible or broken into smaller multiple models leading to sub-optimality. It is now possible to execute these huge optimization models, unlocking enormous potential.

Store-Level Cost-to-Serve Analysis: Previously, calculating the true marginal cost of serving individual stores required running 50-100 scenarios—a week’s-long project that might happen quarterly at best. Now, this analysis can be completed within hours, providing granular insights for account-level decision making.

Daily Route Optimization: Instead of weekly or monthly optimization cycles, supply chain teams can now optimize routes daily, enabling dynamic adjustments to changing market conditions.

Multi-Echelon Scenario Planning: Teams can now model complex what-if scenarios across multiple supply chain tiers simultaneously, evaluating the true end-to-end impact of potential network changes and testing multiple long-term strategies.

As Ganesh explains:

"This isn't merely accelerating existing processes—it's enabling entirely new analytical approaches which build on top of these lightning-fast evaluations. Supply chain leaders are discovering capabilities they hadn't even considered because computational barriers have been removed."

The Road Ahead: Expanding to All of Supply Chain Decision Making

Building on our initial success with network planning and route optimization, Lyric is collaborating with NVIDIA to apply cuOpt GPU acceleration to broader supply chain decision-making problems as well. This initiative aims to revolutionize how companies design, plan and operate their supply chain end-to-end.

As the first Enterprise AI platform dedicated exclusively to supply chain functions, Lyric is committed to democratizing AI-driven decision intelligence. Our no-code application interface ensures that manufacturers, distributors, retailers, and logistics providers can deploy these sophisticated optimization and planning solutions without needing large in-house data science teams.

In today's hyper-competitive landscape, where supply chain agility and efficiency define market leaders, Lyric and NVIDIA cuOpt offers enterprises a transformative capability—one that converts computational power into tangible business outcomes and sustainable competitive advantage.