Perplexity Taps Crusoe for Full AI Model Lifecycle Infrastructure
Event summary
- Perplexity will run its full AI model lifecycle on Crusoe Cloud, including training and inference.
- Crusoe will deploy NVIDIA GB300 NVL72-powered clusters and InfiniBand Solutions for Perplexity's workloads.
- Crusoe will adopt Perplexity Enterprise Pro and Max for its 1,800 employees.
- The partnership is a multi-year agreement, underscoring long-term strategic alignment.
The big picture
This partnership highlights the growing need for specialized AI infrastructure that can support the entire model lifecycle, from training to production. It also underscores the strategic importance of low-latency, high-throughput inference for real-time AI applications. The deal reflects a broader industry shift toward vertical integration in AI, where companies seek to control their entire tech stack for better performance and cost efficiency.
What we're watching
- Scale and Performance
- How Crusoe's infrastructure will handle Perplexity's real-time AI workloads and whether it can maintain low-latency, high-throughput performance at scale.
- Strategic Alignment
- Whether this partnership signals a broader trend of AI companies vertically integrating their infrastructure needs.
- Competitive Dynamics
- The pace at which other AI infrastructure providers respond to this partnership, potentially leading to similar long-term deals.
Related topics