Achieving Real-Time Generative AI During Gameplay for Games on Kubernetes

Explore the integration of real-time generative AI in games running on Kubernetes in this insightful conference talk. Delve into the challenges and solutions for achieving sub-second latency in GenAI applications during gameplay. Learn about deployment options for GenAI inference alongside Kubernetes applications, focusing on low latency, high performance, fast scalability, and cost-effectiveness. Witness a demonstration of real-time GenAI inference integrated with distributed game servers using OSS Agones. Discover how solving this demanding use case can benefit the entire Kubernetes ecosystem and revolutionize game development through applications like creating art assets, auto-generating game code, enabling life-like conversations with bots, and generating levels from player input.