AMD Fell on Monday. The Numbers Tell a Different Story.

AMD closed down 1.16% on June 1. On its own, not worth much attention. But the reason behind it is — and honestly, I think the market got this one pretty wrong.

Nvidia took the stage at Computex 2026 and announced RTX Spark, an Arm-based PC superchip built with Microsoft. It targets premium Windows laptops and compact desktops. AMD gapped down around 5% at the open, recovered most of it, and closed off 1.16%. Intel fell harder. The fear makes surface-level sense: Nvidia is now trying to enter a market Intel and AMD have owned for decades. That is not nothing.

But here is the thing — AMD does not really live in that market anymore. Not primarily.

Q1 2026 revenue was $10.3 billion. Up 38% year-over-year. Ahead of the $9.9 billion Wall Street was expecting. The data center segment alone brought in $5.8 billion — up 57% from a year ago — driven by EPYC CPU and Instinct GPU demand. Non-GAAP gross margin came in at 55%, up 170 basis points year-over-year as the product mix tilts toward higher-value accelerators. Free cash flow hit a record $2.6 billion for the quarter. Q2 guidance is approximately $11.2 billion, which at the midpoint implies roughly 46% growth year-over-year. Management also said lead customer forecasts for the MI450 and Helios rack-scale platform are already running ahead of internal plans. That last part deserves more attention than it got.

Slight tangent, but it matters: the Meta relationship is not what it was six months ago. In February 2026, AMD and Meta announced a multi-year, multi-generation agreement covering up to 6 gigawatts of AMD Instinct GPU deployments. The first 1GW tranche is expected to ship in the second half of 2026, built on a custom MI450-based GPU running on the Helios rack-scale architecture. Meta is also a lead customer for 6th Gen AMD EPYC CPUs. That is not a pilot program. That is a committed infrastructure buildout across multiple product generations.

AWS, Google Cloud, Microsoft Azure, and Tencent have all expanded their EPYC-powered cloud instances. The MI300X — the generation before MI450 — is already running Llama inference workloads at scale inside Azure and Meta. The pipeline behind it is under contract. Multi-year in scope.

So what was Monday actually about.

RTX Spark is real. Microsoft, Dell, HP, ASUS, Lenovo, and MSI are all OEM partners for a fall 2026 launch. The premium Windows PC segment is a place AMD competes. That competitive pressure is legitimate and I would not wave it off. But Windows-on-Arm still carries real friction — app compatibility gaps, DRM issues, and a developer ecosystem that has not fully committed. x86 does not have those problems. And AMD’s client CPU business, even if it faces pressure, generated $2.9 billion in Q1. It is not the growth engine. The data center is.

The part people skipped on Monday: Barclays raised its AMD price target to $665 that same day. Overweight rating intact. More than 20 brokerages lifted targets after the Q1 beat. The analyst community was not reading the Computex headlines the same way the intraday move implied.

AMD CEO Lisa Su described Q1 as “a clear inflection in our growth trajectory and a structural shift in our business.” AMD has also revised its server CPU market growth forecast upward — from 18% to 35% annually — and now projects that market could exceed $120 billion by 2030. Whether you believe that number or not, the direction of the revision matters.

The real question heading into the back half of 2026 is MI450 execution. That is where the thesis either holds or it does not. RTX Spark laptops are a footnote.