Crusoe Expands NVIDIA Collaboration Across the Full AI Factory Stack, Delivering the Complete Infrastructure for the Agentic AI Era

‍Crusoe announces launch support for NVIDIA Vera CPU and NVIDIA Nemotron 3 Super and NVIDIA Nemotron 3 VoiceChat, adds support for NVIDIA Omniverse DSX Blueprint, and contributes Crusoe Managed Inference tokenizer to NVIDIA Dynamo open source framework

NVIDIA GTC, San Jose, CA – March 16, 2026 – Crusoe, the industry’s first vertically integrated AI infrastructure provider, today announced a deepened strategic collaboration with NVIDIA spanning models, inference, and physical infrastructure. The announcements, made at NVIDIA GTC 2026, solidify Crusoe’s position as a full-stack, NVIDIA-native AI infrastructure provider purpose-built for the agentic AI era.

“We’re building every layer of Crusoe's stack in lock step with NVIDIA's roadmap in order to manufacture intelligence with unprecedented speed and scale," said Chase Lochmiller, Co-Founder and CEO at Crusoe. "We’re aligning at every level – from the physical design of our AI factories to how we support their latest models and hardware. This tight integration is how we continue to deliver the exceptional infrastructure solutions our customers require. Crusoe is proud to work with NVIDIA to power the future of agentic AI.”

NVIDIA Vera CPU: Next-Generation Compute for Agentic AI

Crusoe will be an early adopter of NVIDIA Vera CPU, and will be deployed into Crusoe Cloud alongside its NVIDIA Rubin GPU and NVIDIA Vera Rubin NLV72 deployments targeted for late 2026 and throughout 2027. Purpose-built for the demands of agentic AI post-training, where CPU sandbox environments must keep pace with GPUs at every step, a single Vera CPU Rack supports over 22,500 concurrent environments — keeping accelerators fully utilized at AI factory scale.

The Infrastructure Foundation: NVIDIA Omniverse DSX Blueprint

Crusoe will adopt the Vera Rubin DSX Reference Design and the best practices outlined in the NVIDIA Omniverse DSX Blueprint to inform the design and operation of its next-generation gigawatt-scale AI factories. In the future, Crusoe will leverage digital twins for layouts and visualizations, AI-driven power and cooling optimization, and future mechanical and electrical integrations — giving customers infrastructure that is redundancy-designed, energy-optimized, and future-proofed for the evolution of chip technology.

Powering the Full Inference Stack: NVIDIA Nemotron 3 Super, NVIDIA Nemotron 3 VoiceChat, and NVIDIA Dynamo

Crusoe is an early adopter of NVIDIA Nemotron 3 Super and NVIDIA Nemotron 3 VoiceChat, which are now available in the Crusoe Intelligence Foundry and delivered via the Crusoe Managed Inference service powered by MemoryAlloy™ technology.

Nemotron 3 Super is a hybrid MoE model delivering over 50% higher token generation than leading open models today, with support for up to one million token context windows — purpose-built for multi-agent, long-horizon reasoning at enterprise scale. NVIDIA Nemotron 3 VoiceChat is a speech-to-speech model built to support simultaneous speaking and listening. While many current voice agents rely on "cascaded" architectures that stitch multiple components together, this model represents a shift toward a more integrated approach that improves conversational dynamics and fluidity. Both model families are available through the Crusoe Managed Inference Service, which achieves up to 9.9x faster time-to-first-token and 5x higher throughput than standard inference engines. To further accelerate performance, Crusoe is integrating its proprietary tokenizer with NVIDIA Dynamo open-source inference serving library and contributing it back to the community — this high-performance Rust BPE tokenizer delivers approximately 9 times average speedup over HuggingFace tokenizers and up to 31 times faster speedup on long prompts. In agentic workloads, this leads to up to 40% faster time-to-first-token.

"As an early adopter of NVIDIA Nemotron 3 Super and NVIDIA Nemotron 3 VoiceChat, and with the Crusoe Managed Inference tokenizer integration available in NVIDIA Dynamo and open to the community, we're giving AI-native developers everything they need to move from model to production without compromise," said Erwan Menard, Senior Vice President of Product for Crusoe Cloud. "Open models, open infrastructure, and performance that scales — that's the architecture of immediacy in practice."

About Crusoe

As the AI factory company, Crusoe is on a mission to accelerate the abundance of energy and intelligence. The company provides a reliable, scalable, cost-effective, energy-first solution for AI infrastructure. By harnessing large-scale energy sources, building AI-optimized data centers, and delivering a powerful AI cloud platform, Crusoe empowers its customers and partners to build the future faster.