Luxor Energy and Bentaus Complete First AI GPU Response to ERCOT 4CP Signals

Luxor Energy and Bentaus's Ziani™ platform cut a live GPU's power draw to 25% in under 500 milliseconds with no disruption to inference workloads.

HOUSTON, Aug. 24, 2026 /PRNewswire/ -- Luxor Energy, the energy division of Luxor Technology Corporation, and Bentaus, developer of the Ziani™ power asset orchestration platform, today announced that a GPU used for AI inferencing has, for the first time, responded to Four Coincident Peak (4CP) curtailment signals, as part of the Texas ERCOT program. Using signals generated by Luxor's Qualified Scheduling Entity (QSE), Ziani™ software curtailed the GPU in a fully closed loop in under half a second, from signal to verified curtailment, with no disruption to its AI inference workloads.

Power is the binding constraint on AI: data centers are being built faster than grids can add generation and transmission, and inflexible gigawatt-scale loads risk driving up costs and destabilizing the system. By participating in grid programs and curtailing when needed, AI data centers transform from the grid's biggest problem into one of its most valuable resources, unlocking faster interconnection and a more reliable grid.

Participating in these energy programs will be a new revenue source and cost savings for data centers, and will increasingly become more important as compute production commoditizes. ERCOT's 4CP framework allocates transmission costs based on data centers' consumption during the four highest system-wide 15-minute demand peaks of June through September. Large loads that curtail during those intervals can materially reduce their annual transmission charges. ERCOT and other electricity grids have additional programs to financially incentivize participation.

Because the GPU inference workloads can checkpoint, pause, and resume, power consumption can be reduced to approximately 25% of normal operating levels using Ziani software. The software safely ramps down power consumption, while not disrupting its workloads.

"Two trends are converging in the market: AI data centers are consuming a rapidly growing share of total power, while the industry itself is maturing. In that future, the winners will be the data centers that act as grid stabilizers, and get paid for it through energy programs. We will continue to push innovation in this space and tailor it to the nuances of token factories and compute offtake markets," said Ethan Vera, COO of Luxor.

"We are currently extending this work across additional liquid-cooled accelerator systems and next-generation GPU platforms, while developing toward rack-scale architectures. Working with innovative partners like Luxor Energy allows us to connect real-time grid signals directly to compute and prove that AI data centers can become flexible loads that increase grid resilience while creating an entirely new economic opportunity for operators. Together, we've built an end-to-end solution that is much needed for data centers across Texas," said Robert Davidoff, CEO of Bentaus.

The technology has been built and tested, and the two companies are now moving to enroll it in additional grid programs — such as ERCOT's Emergency Response Service (ERS) — running on multiple generations of AI GPUs. R&D is focused on token generation and inference workloads and understanding the profitability impact of enrolling clusters in the energy programs.

About Luxor Energy

Luxor Energy is the energy division of Luxor Technology Corporation, providing retail electricity, QSE services, demand response and dispatch signals, and intelligent load optimization to Bitcoin miners and AI data centers. Luxor Technology Corporation delivers hardware, software, and financial services powering the global compute and energy industry. Learn more at luxor.tech.

About Bentaus

Bentaus designs, deploys, and operates intelligent digital infrastructure for the AI era. Its Ziani™ platform unifies power markets, grid telemetry, and compute infrastructure, transforming AI data centers into grid-interactive resources that respond to energy conditions in real time while maintaining service-level agreements. Learn more at bentaus.com.

Media Contacts

Luxor Technology — Juan Falco, Marketing Analyst [email protected], www.luxor.tech

Bentaus — Matt Stewart, CRO — [email protected], www.bentaus.com

SOURCE Luxor Technology