March 24, 2026
Comino attended NVIDIA GTC 2026 in San Jose this March, joining more than 30,000 people from over 190 countries at the world’s leading AI infrastructure event. NVIDIA CEO Jensen Huang’s keynote made the message clear: the era of inference has arrived. AI is no longer just about experimental training runs; it is now being deployed at scale. Every enterprise, industry, and government is running inference workloads at speeds that are changing how compute infrastructure is built, cooled, and delivered. For Comino, this is exactly what our products were made for.
One of the highlights of our week at GTC was meeting with the NVIDIA leadership team. These discussions showed that our partnership is growing stronger as we move further into the Blackwell generation. Comino now offers the NVIDIA DGX Spark along with our flagship Grando liquid-cooled servers, which use H200 and RTX PRO 6000 GPUs. This means we cover the full range of AI computing, from personal AI supercomputers to multi-GPU inference clusters. The NVIDIA team was engaged and supportive, and sees Comino as a key partner in bringing liquid-cooled AI infrastructure to market.
GTC is known for speeding up partnerships, and this year was no different. We had valuable talks with PNY, one of NVIDIA’s top board partners, about growing our collaboration in EMEA. We also continued to strengthen our relationship with Puget Systems, our US channel partner, which has played a key role in bringing Comino’s liquid-cooled servers to American customers across fields such as AI research, defense, and entertainment.
In addition to our current partners, GTC helped us connect with major companies like TD Synnex and several international firms interested in OEM and white-label opportunities with Comino. The feedback was clear: organizations need dense, quiet, and thermally self-sufficient GPU servers that can be used anywhere. That is exactly what Grando provides.
The trends we saw at GTC 2026 point to strong growth ahead for Comino. The global data center liquid-cooling market is expected to reach $0.6 billion this year, growing at nearly 29% each year. Goldman Sachs predicts that 76% of AI servers will need liquid cooling by the end of 2026, up from just 15% in 2024. This shift is already helping our business grow. Our partnership with NVIDIA is getting stronger, our channel with Puget Systems is bringing in customers like Lockheed Martin and leading AI startups, and we are building new integrator and OEM relationships at GTC. As we move into the second half of 2026, Comino has a stronger pipeline, wider market access, and a product line from DGX Spark to multi-GPU Grando clusters that is ready for the inference era. The turning point is here. We are not just ready for it; we helped build the infrastructure that makes it possible.
