Search

OpenAI Signs 457MW AI Compute Deal with TensorWave Backed by AMD GPUs

By: IDCNOVARegion: North America
OpenAI has signed a six-year agreement with TensorWave, an AMD-focused neocloud provider, for 457MW of dedicated AI compute capacity, marking one of the largest single AI infrastructure contracts to date and deepening the partnership between OpenAI and chipmaker AMD.

The deal will see TensorWave deliver IT capacity powered by AMD Instinct MI455X GPUs across multiple data center locations, operated as dedicated single-tenant infrastructure for OpenAI. The capacity will be built on AMD's Helios rack-scale systems, with operations expected to begin in the first half of 2027 and full deployment targeted by early 2028. The agreement represents TensorWave's largest deployment contract to date and signals a new phase in AI infrastructure procurement, where access to hundreds of megawatts of advanced compute is becoming a decisive competitive factor.

The Helios racks, launched in July, integrate AMD Instinct MI455X GPUs, 6th Gen AMD Epyc "Venice" server CPUs, AMD ROCm software, and AMD Pensando networking into a unified rack-scale architecture. According to AMD, a single rack delivers up to 2.9 exaflops of peak FP4 performance, 1.4 exaflops of peak FP8, 31 terabytes of HBM4 memory, and 1.7 petabytes per second of memory bandwidth. The company claims its Helios platform offers 15 percent more peak FP4 performance, 50 percent more high-bandwidth memory capacity, six percent more HBM bandwidth, and 50 percent more scale-out bandwidth compared with Nvidia's Vera Rubin NVL72 rack, while delivering up to 30 percent more tokens per dollar.

"AI infrastructure is entering a new phase, where the ability to deliver hundreds of megawatts of advanced compute reliably and at speed will separate the companies that can support frontier AI from those that cannot," said Piotr Tomasik, president and co-founder of TensorWave. "TensorWave was built to prove that AMD infrastructure could operate at the scale and reliability that the world's most demanding AI companies require. This deployment takes that work to an entirely new level."

Uday Ruddarraju, CTO for compute at OpenAI, said the agreement advances the company's previously announced partnership with AMD. "As demand for AI continues to grow, expanding access to reliable, large-scale compute is critical to our ability to build and deliver increasingly capable models," he said. "We're excited to advance our previously announced partnership with AMD through this agreement with TensorWave, which will help us continue scaling AI to serve more people and organizations around the world."

The deal underscores the growing role of neocloud providers in supplying large-scale AI capacity and highlights AMD's push to challenge Nvidia's dominance in AI accelerators. TensorWave, which secured $350 million in Series B funding in June 2026, claims to have secured 2GW of long-term data center capacity across the United States. The company is a known customer of TecFusions, having signed an additional 20MW capacity agreement in January 2026 for data centers in Pennsylvania and Arizona. For OpenAI, the contract represents a significant step in diversifying its compute supply chain beyond Nvidia and securing the infrastructure needed to support increasingly capable frontier models.