
WhiteFiber (NASDAQ:WYFI) reported first-quarter revenue growth and positive adjusted EBITDA as management said the company is nearing initial capacity delivery at its NC1 data center project in North Carolina and continuing to reposition its cloud business toward longer-duration enterprise deployments.
On the company’s first-quarter 2026 earnings call, Chief Executive Officer Sam Tabar said White Fiber delivered “year-over-year revenue growth, strong gross margins, and positive adjusted EBITDA” while investing in its AI infrastructure platform. He said demand for AI infrastructure remains strong, but emphasized that execution — including power, equipment, capital, customer requirements and construction — is the primary constraint in the market.
NC1 Nears Initial Capacity Delivery
Tabar said Duke Energy has completed delivery of the initial 54 gross megawatts of utility power to NC1, supporting the first 40 megawatts of IT load under White Fiber’s agreement with Nscale. He said construction and commissioning remain active, with approximately 600 personnel on site during the prior week.
Major equipment packages, including generators, UPS systems and chillers, are on site, Tabar said. The company is managing a remaining supply chain issue involving certain medium-voltage switchgear components. He said White Fiber expects to begin delivering initial capacity “within the next couple of weeks,” though delivery may begin slightly later than June 1.
“Based on our current discussions, we do not expect a material delay, a material impact to our customer’s commissioning process, or a change to the overall project economics,” Tabar said.
NC1 is backed by a long-term agreement with Nscale, which Tabar said is supported by Nscale’s investment-grade hyperscaler offtake. He added that White Fiber expects to begin marketing the next 45-megawatt tranche of capacity this summer and is working with the utility on longer-term power expansion opportunities that could scale the site to approximately 300 gross megawatts over time.
In response to an analyst question, Billy Krassakopoulos, White Fiber’s President of Data Centers, said the initial 54 megawatts are fully contracted with Nscale. He said the additional 45 megawatts are expected to deliver between the middle and end of 2027 and would be marketed after phase one is delivered, with Nscale expected to receive the first opportunity for that capacity.
MTL-3 Now Owned, Supporting Cerebras
Tabar said the first quarter was the first full quarter of operations for MTL-3, which supports White Fiber’s colocation agreement with Cerebras. After the quarter ended, White Fiber completed the purchase of MTL-3 through a previously disclosed purchase option, supported by an amended credit facility from the Royal Bank of Canada.
Tabar said owning the site is expected to reduce lease payments by approximately CAD 3.1 million, or $2.3 million, annually over the remaining term. He added that White Fiber has submitted an application with the local utility to more than triple the site’s available power over time, subject to review and approval.
Krassakopoulos said in the Q&A session that power is the key gating item for any MTL-3 expansion. He said indications from the utility are positive, but the company does not yet have timing information. If additional power is approved, he said Cerebras would be “one of the first customers” White Fiber speaks with.
Cloud Business Repositioning Continues
Tabar said White Fiber’s cloud business is undergoing a strategic shift toward longer-duration enterprise deployments, managed infrastructure services and next-generation GPU capacity. The transition has created near-term revenue pressure, and management continues to expect the second quarter to be the low point for cloud revenue.
Tabar said the company is in the final stages of a long-duration, nine-figure cloud opportunity with a high-quality enterprise customer in a new market. He said White Fiber would provide more detail if the agreement is executed and customer disclosure approvals are in place.
The company also signed a two-year agreement with Hyperbolic for approximately CAD 17 million of total contract value, supporting Modo Labs as the end customer. Tabar said the deployment will use H200 GPUs from White Fiber’s existing owned fleet as part of a cross-data-center research and development project and will not require incremental GPU capital expenditures.
Tabar said White Fiber is tracking more than 50,000 GPUs in its cloud pipeline, representing a weighted pipeline value of approximately $3.3 billion based on stage and probability of close. He cautioned that not all of the pipeline will convert and said the company will remain selective.
During the Q&A session, Billy Kladeck, Head of Revenue, said most incremental cloud opportunities under review involve GPU generations beyond H200s, “mostly the B300s.” He said the company is seeking minimum three-year terms, cash-flow-positive economics over the life of deals and prepayments that reduce the need to use corporate balance sheet capital.
First-Quarter Results and Balance Sheet
Chief Financial Officer Erke Huang said first-quarter revenue was $21.9 million, up 31% from $16.8 million in the first quarter of 2025. Cloud revenue was $16.8 million, compared with $14.8 million a year earlier, while colocation services revenue rose to $4.8 million from $1.6 million, mainly due to MTL-3 beginning billing in October 2025 under the Cerebras agreement.
Gross profit excluding depreciation and amortization was $13.2 million, with gross margin of 60.2%. Huang said depreciation and amortization expense was CAD 6.4 million, compared with CAD 3.8 million in the prior-year quarter, reflecting expansion of cloud and colocation infrastructure.
General and administrative expense increased to CAD 17.8 million from CAD 4.2 million a year earlier, which Huang attributed to share-based compensation, higher headcount, standalone public company costs and platform investment. He said the company expects second-quarter G&A to decrease by around 20% compared with the first quarter.
- Operating loss was $11 million, compared with operating income of $2 million in the first quarter of 2025.
- Interest expense was $2 million, primarily related to convertible notes issued during the quarter.
- Net loss was $12 million, compared with net income of $1.4 million a year earlier.
- Adjusted EBITDA was $3 million, compared with $6 million in the first quarter of 2025.
White Fiber ended the quarter with $75.8 million of cash and cash equivalents and $4.3 million of restricted cash. Huang said the company completed a $230 million private placement of 4.5% convertible senior notes due 2031 and entered into a zero-strike call structure intended to reduce potential dilution.
Management Points to Pipeline, Financing
Tabar said White Fiber is in a formal project-level financing process for NC1, with lender diligence active. He said a financing solution could validate the asset’s credit quality, recycle invested capital and provide flexibility to pursue additional customer-backed opportunities.
Management said the broader pipeline remains active. Tabar said White Fiber is in advanced discussions with two large customers for two distinct site opportunities, including customers with scaled AI infrastructure needs and strong credit profiles. He also said the company expects to close at least one additional site in the coming months, subject to diligence, documentation, customer alignment and capital availability.
Krassakopoulos said White Fiber is focused on grid power for potential site acquisitions and is not pursuing self-generation or off-grid solutions. He said the company’s model is to find sites with power available on day one, bring an initial phase online quickly and then return for later phases as additional power becomes available.
Tabar closed by saying White Fiber’s focus is execution as NC1 nears initial revenue, financing advances, MTL-3 operates under company ownership and the cloud business moves from transition toward renewed growth.
About WhiteFiber (NASDAQ:WYFI)
We believe we are a leading provider of artificial intelligence (“AI”) infrastructure solutions. We own high-performance computing (“HPC”) data centers and provide cloud-based HPC graphics processing units (“GPU”) services, which we term cloud services, for customers such as AI application and machine learning (“ML”) developers (the “HPC Business”). Our Tier-3 data centers provide hosting and colocation services. Our cloud services support generative AI workstreams, especially training and inference.
