SharonAI Holdings, Inc. Class A Common Stock (NASDAQ:SHAZ) said it expanded its secured AI factory capacity, contract backlog and funding during the second quarter, as the company works to deploy GPU infrastructure across Australia and New Zealand.
Chief Executive Officer and co-founder James Manning said the company had secured 212 megawatts of total AI factory capacity as of the call date, up 80 MW from its prior guidance of 132 MW. Of that total, 120 MW has been contracted under multiyear take-or-pay agreements, while 92 MW remains available to sell. SharonAI expects to deploy more than 64,000 NVIDIA GPUs by mid-2027.
Contracted Revenue Book Reaches $8.8 Billion
SharonAI said it has executed approximately $8.8 billion in total contract value year to date. Manning said the company’s portfolio had consisted of 100 MW of capacity and $2.2 billion in total contract value three months earlier, meaning secured capacity has more than doubled while the contracted book has increased roughly fourfold.
The company highlighted several recently announced agreements:
- A six-year strategic compute collaboration with NVIDIA with $4.9 billion in minimum revenue, involving an initial 72 MW deployment and 40,000 GB300 GPUs.
- A five-year take-or-pay agreement with a global AI lab valued at $1.32 billion in total contract value.
- A five-year take-or-pay agreement with a global technology company valued at $950 million.
- A five-year take-or-pay agreement with a global AI platform valued at $373 million.
Manning said the most recent global AI platform agreement involves a B300 deployment priced at more than $4 per GPU hour, which he described as a record rate for the company. He said demand for GPU capacity remains constrained and that SharonAI has been able to pursue higher pricing in customer discussions.
Under the NVIDIA arrangement, NVIDIA provides a six-year anchor commitment intended to support financing for GPU and related infrastructure investments. Manning said the agreement establishes a minimum price rather than a ceiling. SharonAI expects to contract a significant portion of the capacity at prices above the minimum and retain the anchor price while sharing incremental revenue with NVIDIA.
Capacity Expansion and Deployment Plans
The additional 80 MW of capacity announced during the quarter comes from a new partner and is located in Australia, according to Manning. He said some initial megawatts could become available as early as late 2026, with further delivery expected through 2027.
SharonAI began the year with 54 MW of secured capacity, according to Manning. The company has also established its first New Zealand facility, supporting its sovereign infrastructure strategy for customers with data residency and regulatory requirements.
The company said it has handed over a B300 cluster to a customer and is working on further B300 and GB300 deployments through the remainder of the third quarter and into early fourth quarter. Manning said SharonAI expects its first material revenue to begin in the fourth quarter of 2026 as large-scale deployments come online.
During the question-and-answer session, Manning said the 40,000 GPUs included in the NVIDIA-backed program are expected to be deployed during the first half of 2027. He said the company expects the vast majority of that capacity to be sold under three- to five-year take-or-pay contracts, primarily to AI-native companies such as model builders and inference providers. A smaller portion may be available for spot demand.
Management said the primary factors that could affect the revenue ramp are hardware delivery and data-center readiness. Manning said the company is tracking supplier deliveries and its data-center partners’ construction and readiness milestones.
Storage and Capital Strategy
SharonAI expanded its storage commitment with VAST Data to 600 petabytes, which Manning said could support growth to as many as 100,000 GPUs. He also said customer requirements are increasing storage needs per GPU cluster. One recent contract, for example, includes 10 PB of storage per 1,000 GPUs, compared with a standard reference architecture of about 3 PB per 1,000 GPUs.
The company said it has raised approximately $2.2 billion since December 2025, including a $1.6 billion oversubscribed strategic financing round completed in June and a $350 million convertible note completed in April. Manning said SharonAI is also advanced in discussions concerning debt facilities, though he did not provide details on prospective financing terms.
SharonAI said its capital-efficient model relies on partnerships with data-center operators rather than developing greenfield facilities. The company identified NVIDIA as its primary compute supplier, NEXTDC as its primary provider of data-center capacity, VAST Data as a storage partner, and World Wide Technology as its exclusive APAC procurement, testing and implementation partner.
Leadership Changes
The company announced several leadership appointments, including Anuj Goel, formerly of Macquarie Group, as chief financial officer; Melissa Anastasiou as chief legal officer; and Andrew Penn as non-executive chairman of the board.
Manning also thanked outgoing CFO Tim Broadfoot, while Ross Barrows, head of capital strategy and investor relations, said the company’s latest quarterly filing would be Broadfoot’s final 10-Q. Broadfoot will continue consulting with SharonAI for a period, Barrows said.
Looking ahead, Manning said SharonAI’s contracting visibility extends through 2031 and that the company plans to focus on execution, delivery and operating performance over the next 12 months. “2026 and 2027 will be the year about delivery for us,” he said.
About SharonAI Holdings, Inc. Class A Common Stock (NASDAQ:SHAZ)
SharonAI Holdings Inc is a high-performance computing (HPC) company deploying large-scale energy and compute infrastructure, USA energy markets and infrastructure asset management. Its services include: Sovereign AI Australia, GPU-as-a-Service, SHARON AI Cloud, SHARON AI Private Cloud, Virtual Private Clusters, HPC Servers, SHARON AI Supercluster, GPU Fleet, Virtual Servers, Cloud Storage, AI Model Training, High-Performance Computing (HPC), and Video Encoding & Decoding. The company’s products are: Sovereign AI Australia, GPU-as-a-Service, SHARON AI Cloud, SHARON AI Private Cloud, Virtual Private Clusters, HPC Servers, SHARON AI Supercluster, GPU Fleet, Virtual Servers, Cloud Storage, AI Model Training, High Performance Computing (HPC), and Video Encoding & Decoding.
