OpenAI Signs 457MW Deal With AMD-Powered Cloud Provider TensorWave
TensorWave, an AMD-powered cloud provider, signed a six-year deal with OpenAI to build 457MW of AI computing capacity using AMD's Instinct MI455X processors. The agreement reserves all capacity for OpenAI, with deployment starting in 2027. AMD Ventures co-led TensorWave's $350M funding round, valuing it at $1.55B. The deal highlights competition in AI chip market ahead of Nvidia's earnings report.
How this was made

The 30-second read
Why it matters
The deal strengthens AMD’s competitive positioning in AI infrastructure by tying its newest GPU platform to a major AI developer’s dedicated capacity, while also setting up a near-term narrative contrast with Nvidia’s upcoming earnings.
Market read
Traders may reassess AI GPU competitive dynamics as AMD secures a large, dedicated compute allocation for OpenAI, though financial magnitude and timing are not fully specified.
What to watch
Dedicated 457MW capacity does not guarantee AMD’s share of total AI GPU spend, and the first capacity is not expected until H1 2027, limiting immediate earnings impact.
Background
TensorWave, a Las Vegas-based AI cloud provider, signed a six-year agreement to build 457MW of OpenAI-dedicated compute using AMD Instinct MI455X and Helios rack-scale technology.
Ticker impact
OpenAI’s six-year compute deal is set to run entirely on AMD Instinct MI455X, reinforcing AMD’s push to win AI infrastructure share.
Shares may see supportive sentiment as the market prices in incremental demand and credibility for AMD’s latest Instinct platform.
The article ties a large 457MW OpenAI allocation to AMD MI455X and to a broader OpenAI-AMD partnership, but it does not disclose deal value or AMD financial impact.
The TensorWave announcement is timed a day before Nvidia’s quarterly results, potentially shaping expectations for AI chip demand and competitive dynamics.
Likely limited immediate impact, but could add competitive pressure to Nvidia’s narrative ahead of its earnings.
The article does not state Nvidia’s results or any direct customer loss, only that Nvidia is scheduled to report soon.
Market effects
Reinforces the AI cloud infrastructure shift from Nvidia-centric builds toward multi-vendor GPU supply, potentially affecting how investors underwrite AI compute demand.
US data-center buildout focus could influence regional power, construction, and data-center capacity expectations, though locations are not disclosed.
Signals continued scaling of AI compute capacity and intensifying GPU competition globally as major developers secure dedicated infrastructure.
Counterpoint
Without disclosed deal value, power-site locations, or confirmed MI455X ramp details, the market may treat this as strategic validation rather than near-term revenue upside.
Key entities
- customerOpenAI
Major AI developer reserving all 457MW of compute capacity under the agreement.
- cloud_providerTensorWave
AI cloud provider building the 457MW infrastructure for OpenAI across US data-center sites.
- chip_supplierAMD
Provides Instinct MI455X GPUs and is linked to a broader six-gigawatt OpenAI strategic partnership.
- peerNvidia
Scheduled to report quarterly results the day after the TensorWave announcement, framing competitive expectations.



