Redwood AI Upgrades Platform for Faster Studies and CPU Deployment
Redwood AI has released a performance upgrade for its AI-powered synthesis prediction model. The upgrade enables faster studies and broader CPU-based deployment, reducing reliance on specialized hardware.
Vancouver, Canada – Redwood AI Corp. announced a significant performance upgrade to its proprietary AI-powered chemical synthesis prediction model. The upgrade is designed to enhance model efficiency and runtime speed, allowing for quicker full studies and expanding deployment flexibility beyond specialized hardware.
A key benefit of the upgrade is its ability to run entire studies on standard CPU infrastructure, a shift from previous reliance on GPU hardware. This aims to broaden accessibility for organizations that need powerful AI models without restrictive cloud setups or in environments with GPU procurement delays, such as air-gapped systems.
The company states the upgrade aligns with trends in "AI sovereignty," supporting biopharmaceutical organizations that prioritize keeping compute and sensitive workflows within their own infrastructure. By enabling effective CPU execution, Redwood AI suggests users can deploy closer to where data resides, whether on-premise or in controlled environments, while maintaining performance.
Redwood AI also anticipates that the improved efficiency and reduced infrastructure requirements will lower ongoing operational and hosting costs. This is particularly beneficial for organizations scaling usage across teams or projects where predictable compute budgets and consistent performance are crucial.
"We built this performance upgrade to make the platform more dependable under real-world constraints," said Louis Dron, CEO of Redwood AI. "Users should not have to choose between speed, control, and cost. By improving efficiency and enabling full studies to run on CPUs, teams can keep workflows moving even when infrastructure is constrained."