Micron Technology is having the kind of year most companies only dream about. Revenue is up 346% year-over-year, gross margins sit around 85%, and every chip the company can produce is already spoken for. Yet even a business expanding at this pace is not immune to the product schedule of its biggest customer. Nvidia's Vera Rubin platform ramp has created a timing question, and some analysts now expect portions of Micron's revenue to be recorded in fiscal 2027 rather than the current fiscal year.

A revenue machine running at full tilt

Micron's fiscal third quarter of 2026, which ended in May, showed a company riding the AI infrastructure wave with near-perfect form. Revenue reached $41.46 billion for the quarter, and management guided fiscal Q4 to approximately $50 billion as demand continues to surge.

Data center revenue alone exceeded $25 billion in the quarter, putting Micron on an annualized run rate of more than $100 billion from that segment.

The standout product is HBM4, Micron's latest generation of high-bandwidth memory designed specifically for AI accelerators such as Nvidia's upcoming Vera Rubin platform. Micron has already generated more than $1 billion in HBM4 revenue and is ramping production at twice the speed of its predecessor, HBM3E. Because memory generations like HBM4 are introduced in step with new accelerator platforms, the pace of a customer's platform transition feeds directly into a supplier's revenue calendar — the dynamic now in play between the two companies.

CEO Sanjay Mehrotra has signaled confidence by locking down the supply side. Micron has signed 16 strategic customer agreements containing take-or-pay provisions, which obligate customers to either purchase the chips or pay a penalty. Those agreements represent roughly $22 billion in commitments and deposits, and the company carries nearly $100 billion in total remaining performance obligations.

Micron's entire HBM supply for 2026 is completely sold out under multi-year contracts.

The Nvidia timing question

Some analysts now estimate that portions of expected revenue may shift from the current fiscal year into the next, depending on how quickly Nvidia moves through product qualification and volume production phases.

The fiscal-calendar mechanics are worth keeping straight. Micron's fiscal year does not map onto the calendar year — fiscal Q3 2026 ended in May — so fiscal 2027 refers to a later stretch of the company's own reporting schedule. And as analysts frame it, the open question is when the revenue lands, not whether it is contracted: the chips in question fall under the same take-or-pay agreements and the nearly $100 billion in remaining performance obligations already on Micron's books.

The near-term markers are procedural. Nvidia's progression through Vera Rubin qualification and into volume production will set the pacing, and Micron's fiscal Q4 result against the roughly $50 billion guidance will show how much revenue lands inside the current fiscal year.

Supply constraints set to persist

Mehrotra has stated that tight supply conditions for HBM are expected to extend beyond calendar year 2027, with gradual improvement potentially arriving in 2028. Micron is spending aggressively to expand capacity, with capital expenditures projected to exceed $25 billion.

Analysts project approximately 84% revenue growth for Micron in fiscal year 2027, driven largely by this structural imbalance.

The take-or-pay structure of Micron's customer agreements adds another layer of insulation. Even if individual product timelines slip, the company holds contractual protections ensuring it gets paid. Nearly $100 billion in remaining performance obligations represents a revenue floor few semiconductor companies can match.