How they use Nebius AI Cloud
Emerald AI tested its Conductor platform at Nebius’s AI factory outside London during a five-day demonstration. The trial combined its software with a cluster of 96 NVIDIA Blackwell Ultra GPUs to adjust power consumption in response to simulated grid stress.
NVIDIA reports that the platform met more than 200 requested power targets and supported load reductions lasting up to 10 hours. Its separate technical account says simulated high-priority AI workloads maintained peak throughput while more flexible jobs slowed temporarily. That account reports approximately 30% power reduction within 40 seconds during emergency testing.
The demonstration tested grid-responsive operation at a commercial facility, with EPRI and National Grid supplying simulated stress scenarios.
Sources
4 sourcesEmerald AI Showcases How Grid-Flexible AI Factories Unlock Latent Capacity
Listed 10 Nov 2026 · NVIDIA · Web page
NVIDIA describes Emerald Conductor’s five-day demonstration on a 96-GPU cluster at Nebius’s AI factory outside London.
Open original sourceHow Emerald AI Makes AI Factories Power-Flexible
Listed 10 Nov 2026 · NVIDIA · Web page
NVIDIA reports Emerald Conductor’s grid-flexibility demonstration at Nebius’s AI factory outside London.
Open original sourceBlowing Off Steam: How Power-Flexible AI Factories Can Stabilize the Global Energy Grid
Published 25 Mar 2026 · NVIDIA · Web page
Josh Parker’s NVIDIA article reports that Emerald Conductor preserved simulated priority-workload throughput during the Nebius demonstration while slowing flexible jobs. It attributes approximately 30% power reduction within 40 seconds to the follow-on paper.
Open original sourceCompany context
emeraldai.coEmerald AI develops Emerald Conductor, software that helps data center operators adjust AI power use in response to grid conditions. The platform connects data centers with utilities, models expected power demand, coordinates computing workloads and on-site energy resources, and measures performance. Data center developers and operators use it to make their facilities more responsive to grid needs; utilities and grid operators can use it to see and dispatch flexible demand. The company reports deployments and demonstrations with commercial data centers and energy partners.
Recent updates
3 updatesEmerald AI and Lumen explore energy-aware networking collaboration
Emerald AI and Lumen announced a collaboration to develop capabilities for shifting AI computing workloads between locations based on power availability, cost and other factors. They are exploring a pilot using Lumen’s edge and network infrastructure.
Emerald AIEmerald AI, Google and NVIDIA launch the AI Energy Management Alliance
Emerald AI, Google and NVIDIA announced the AI Energy Management Alliance to advance data centers that dynamically manage electricity use in response to grid conditions. The alliance plans technical and operational approaches with utilities and advocacy for policies recognizing grid-responsive demand.
NVIDIA BlogEmerald AI raises $150 million Series A at $1.05 billion valuation
Emerald AI announced a $150 million Series A co-led by Energize Capital and DCVC at a $1.05 billion valuation. The company said its technology was already deployed commercially at multi-megawatt, full data-center scale.
Emerald AI