How an Artificial Intelligence company brought the 1980s text adventure genre to the 21st century with limitless open-ended storylines and affordable economics.
“You are about to enter a world of endless possibilities, where you can do absolutely anything you can imagine…. Will you proceed?”
This is the first message players receive when logging in to AI Dungeon. Leveraging the power of GPT-2 and GPT-3, the company delivers limitless open-ended storylines. The output of their AI language models are more powerful than any deployed at scale to create individualized, real-time experiences for every user.
AI Dungeon has experienced exponential user growth since launch. Initially, AI Dungeon struggled with accessing high-performance GPU compute resources at the speed & scale needed to accommodate demand, and the pricing needed to effectively manage costs. This led to a few problems, including frequent slow downs and outages as they struggled to reliably deploy such massive language models. Together, this created a sub-optimal user experience and higher costs for AI Dungeon.
When AI Dungeon migrated to CoreWeave’s Inference Service, which empowered them to leverage the industry’s fastest auto-scaling, they were able to reduce compute instance spin-up times to 45 seconds and minimize idle compute. For AI Dungeon, this led to less overhead compute, more reliable infrastructure, 76% lower computing costs, and a significant reduction in resource planning. For users, this meant significantly shorter wait times, and a better in-game experience.
“After a few months of struggling to keep up with demand at mega-cloud prices, we were able to seamlessly move our cloud infrastructure over to CoreWeave. Using CoreWeave’s Inference Optimized V100 instances, our inference latencies dropped by 50% and the cost savings from our partnership allowed us to continue delivering our free-tier experience and substantially reduce our cost-per-user.”
- Nick Walton, Founder and CEO of Latitude, the creator of AI Dungeon