Inve US ▾
AMAMD

Advanced Micro Devices (AMD) Q1 FY26 earnings call: guidance and Q&A

Earnings call· Mar 2026 (Q1 FY26)· Information Technology· May 5, 2026

Source: earnings call transcript

Executive summary

AMD Q1 FY26 — Data-center-led inflection as agentic AI drives 38% revenue growth and record FCF

AMD's quarter marks a structural inflection: data center is now the primary earnings driver as agentic AI pulls both accelerator and CPU demand, and management sharply raised its server-CPU TAM view ahead of a second-half Instinct/Helios ramp. The forward stance is aggressively expansionary — leaning into R&D and capacity ahead of demand — tempered by memory-cost pressure weighing on the consumer businesses in the back half.

Highlights

5
  • Revenue $10.3B, up 38% YoY (above the high end of guidance), with every segment growing YoY

  • Data Center revenue a record $5.8B, up 57% YoY and 7% QoQ; segment operating income $1.6B (28% margin) vs 25% a year ago

  • Non-GAAP diluted EPS $1.37, up 43% YoY; non-GAAP gross margin 55%, up 170 bps YoY

  • Record free cash flow of $2.6B (25% of revenue), more than tripling YoY

  • Server CPU revenue up more than 50% YoY (4th consecutive record quarter), with Cloud and Enterprise each up more than 50%

Concerns

5
  • Second-half PC and gaming demand expected to soften on higher memory/component costs; H2 gaming revenue guided to decline more than 20% vs H1

  • Data Center AI revenue down modestly QoQ due to a China revenue step-down (China not material in Q1)

  • Non-GAAP operating expenses up 42% YoY to $3.1B, repeatedly running above prior guidance

  • MI450 ramp in Q4 carries below-corporate-average gross margin, a mix headwind to manage

  • Supply chain, data-center build-out and power described as tight/gating factors into the 2027 ramp

Guidance & targets

CategoryTargetConfidence
Total revenue
~$11.2B, +/- $300M (midpoint +46% YoY, +9% QoQ)
high materiality
High
Non-GAAP gross margin
~56%
medium materiality
High
Non-GAAP operating expenses
~$3.3B
medium materiality
High
Non-GAAP other income and expense
gain of ~$60M
low materiality
High
Non-GAAP effective tax rate
13%
low materiality
High
Diluted share count
~1.66B shares
low materiality
High
Server CPU revenue growth
>70% YoY in Q2, robust growth continuing through H2 2026 and into 2027
high materiality
High
Data Center segment sequential growth
up double digits QoQ, in both Server and Data Center AI
high materiality
High
Server CPU TAM
>35% annual growth, reaching over $120B by 2030
high materiality
Medium
Server CPU share target
greater than 50% market share
high materiality
Medium
Data Center AI revenue
tens of billions of dollars in annual Data Center AI revenue in 2027
high materiality
High
Data Center AI long-term growth
exceed long-term growth target of greater than 80% CAGR in the coming years
high materiality
Medium
Helios / Data Center AI ramp cadence
initial volume in Q3, significant ramp in Q4, continuing to ramp in Q1
high materiality
Medium
Second-half gaming revenue
decline more than 20% vs first half
medium materiality
Medium
Client revenue (full year)
grow year-over-year and outperform the market
medium materiality
Medium
H2 PC shipments
planning for lower second-half PC shipments
medium materiality
Medium
Long-term gross margin
55% to 58% range
medium materiality
Medium
Long-term EPS
more than $20 in EPS over the strategic time frame
high materiality
Medium
Opex mix (R&D vs SG&A)
R&D to grow year-over-year much faster than SG&A for the year
low materiality
Medium

Segment performance

SegmentRevenueYoYQoQMargin
Data Center
Record quarter led by EPYC CPUs and ramping Instinct GPUs; AI was the primary driver of cloud growth via general-purpose compute, data processing, head nodes and agentic applications. Enterprise set record revenue and sell-through with new wins across financial services, healthcare, industrial and digital infrastructure.
Server CPU revenue: >50% YoY growth (4th consecutive record quarter)Cloud revenue growth: >50% YoYEnterprise revenue growth: >50% YoYEPYC-powered cloud instances: >1,600 (+~50% YoY)Data Center AI revenue: significant double-digit % YoY growth, down modestly QoQ on China
$5.8B+57%+7%Operating income $1.6B (28% of revenue) vs $932M (25%) a year ago
Data Center — Server (sub-segment)
Fourth consecutive quarter of record server CPU revenue; share gains accelerated YoY on Turin ramp and continued 4th-gen EPYC strength across cloud and enterprise.
Turin (5th-gen EPYC) crossed >50% of revenueGenoa (Zen 4) still shipping strongly; Milan declining
—>50%——
Data Center — Data Center AI (sub-segment)
Instinct adoption expanding across cloud, enterprise, sovereign and supercomputing; QoQ decline attributed to lower China revenue (China not material in Q1). Q2 guided to double-digit sequential growth.
Largest deployments are for inferenceMI450 series sampling to lead customers
—significant double-digit %down modestlybelow corporate-average gross margin as MI450 ramps
Client and Gaming
Sequential decline consistent with seasonality; H2 demand expected softer on higher memory/component costs.
Client revenue: $2.9B (+26% YoY, -7% QoQ)Gaming revenue: $720M (+11% YoY, -15% QoQ)
$3.6B+23%-9%Operating income $575M (16% of revenue) vs $496M (17%) a year ago
Client and Gaming — Client (sub-segment)
Driven by latest Ryzen processors, favorable mix and share gains across consumer and commercial; Dell, HP and Lenovo broadened AMD offerings; new enterprise wins across technology, financial services, healthcare and aerospace.
Commercial Ryzen Pro PC sell-through: >50% YoYNotebook/mobile (esp. premium) growing; desktop softer
$2.9B+26%-7%—
Client and Gaming — Gaming (sub-segment)
Graphics up YoY on Radeon 9000; semi-custom down at this stage of the console cycle; H2 gaming revenue guided to decline >20% vs H1 on memory/component costs.
Radeon 9000 series demand up YoYSemi-custom (console) revenue declining as expected in the cycle
$720M+11%-15%—
Embedded
Return to growth on strength in test/measurement/emulation, aerospace & defense and communications, plus adoption of embedded x86; portfolio expanding from FPGA-focused to adaptive x86 and semi-custom, growing TAM.
Design-win momentum: +double-digit % YoYBillions of dollars in new design winsAccretive to corporate gross margin
$873M+6%-8%Operating income $338M (39% of revenue) vs $328M (40%) a year ago

AMD operating KPIs by quarter

AMD operating KPIs stated on its earnings calls, by fiscal quarter
KPIThis call Mar 2026 Q1 FY26 Jun 2026 Q2 FY26Change vs prior quarter
EPYC-powered cloud instances
1,600+ EPYC-powered cloud instances increased nearly 50% year-over-year to more than 1,600 with instances optimized for virtually every enterprise workload and expanded availability across the largest global cloud providers. Source: earnings call transcript
1,600+ Fifth-gen EPYC Turin now powers nearly 1/3 of the more than 1,600 EPYC public cloud instance types available globally as providers broaden their offerings with new database storage and AI workloads. Source: earnings call transcript
—

Operating figures the company stated on at least two of the calls shown, checked against each call's transcript. Click a figure to read the sentence. A dash means it was not stated that quarter.

Product announcements

ProductTypeDetails
6th-gen EPYC 'Venice' (Zen 6, 2nm), incl. 'Verano'roadmap
Instinct MI450 series (incl. Meta custom accelerator)milestone
Helios rack-scale AI platformmilestone
Instinct MI355X / MI355milestone
Instinct MI500 seriesroadmap
Ryzen AI 400 series and Ryzen AI Pro 400 series desktop CPUslaunch
Radeon 9000 series GPUs / FSR software updateupdate
ROCm software stackupdate

Deals & partnerships

Meta Strategic partnership / custom accelerator supply up to 6 gigawatts of AMD Instinct GPUs, spanning several product generations multi-generation

Expanded strategic partnership to deploy up to 6 GW of Instinct GPUs, including a custom GPU accelerator based on the MI450 architecture co-designed for Meta's next-gen AI workloads, leveraging Helios rack-scale architecture.

OpenAI Strategic partnership (AI infrastructure) multiyear

Referenced as a previously announced partnership positioning AMD as a core partner to the world's largest AI infrastructure builders; co-engineering going well.

Risks & headwinds

Higher memory and component costs pressuring consumer PC and gaming demand second half of 2026

H2 gaming revenue expected to decline more than 20% vs H1; H2 PC shipments planned lower

Mitigation:Focus on premium/commercial segments and Ryzen share gains; still expect Client revenue to grow YoY and outperform market; pairing every CPU/GPU shipment with secured memory

Memory price inflation / rising input costs current and forward

Unquantified cost increases on the memory side in a tight supply environment

Mitigation:Deep partnerships with memory vendors; secured enough supply to meet and exceed targets; sharing some cost inflation with customers via server ASPs

Supply chain, data-center build-out and power constraints into 2027

Unquantified; described as tight across wafer, back-end capacity, build-outs and power

Mitigation:Deep supply-chain relationships; increasing wafer and back-end capacity; more power coming online in 2027; visibility down to specific data centers

China revenue / export-related transition in Data Center AI Q1 FY26

Data Center AI down modestly QoQ; China revenue not material in Q1 (was higher in Q4)

Mitigation:Broadening customer and workload base outside China; Q2 Data Center AI guided up double digits sequentially

Gross-margin dilution from MI450 ramp H2 FY26 (esp. Q4)

MI450 below corporate-average gross margin, ramping Q3 and significantly in Q4

Mitigation:Multiple tailwinds (server CPU, richer Client mix, lower Gaming, accretive Embedded) to offset; future ASP and cost/scale levers on Instinct

Rising / persistent non-GAAP operating expenses ongoing

Opex +42% YoY to $3.1B; guided ~$3.3B in Q2; repeatedly above prior guidance

Mitigation:Framed as demand-driven investment tied to revenue momentum; R&D to grow faster than SG&A going forward

Competitive intensity in server CPU (x86 competitor supply, ARM merchant/custom) ongoing

Unquantified; competitor improving supply and restarting 7nm; ARM introductions in market

Mitigation:Broad optimized CPU portfolio (Venice/Verano) vs point products; deep hyperscaler co-optimization; >50% share target

Semi-custom (console) cyclical decline Q1 FY26

Semi-custom revenue declined YoY as expected at this stage of the console cycle

Mitigation:Strong engagements with customers on next-generation console platforms

Q&A highlights

Why has the 2030 server CPU TAM roughly doubled so quickly since the Analyst Day, and how confident is AMD in the >50% share target given improving x86 competitor supply and ARM momentum?

Su said the AI-driven CPU pull first seen last year (prompting the ~$60B/18% view) is happening much faster; agentic AI and inferencing require substantial CPU compute for orchestration and data processing, verified bottoms-up via customer forecasts and workload analysis, yielding the >35%/$120B view. She framed CPUs in three buckets and argued a broad portfolio wins, reaffirming confidence in >50% share on Turin traction and Venice positioning.

“we now expect the server CPU TAM to grow at greater than 35% annually, reaching over $120 billion by 2030”

asked by Joshua Buchalter · answered by Lisa Su

3 min read 6 chapters

Detailed narrative

Structural inflection with data center as the primary growth engine

AMD delivered Q1 revenue of $10.3B, up 38% YoY and above the high end of guidance, with every segment growing YoY and earnings up more than 40%. Data Center revenue reached a record $5.8B (up 57% YoY), and management characterized the quarter as a clear inflection and structural shift, with data center now the primary driver of revenue and earnings.

Free cash flow more than tripled to a record $2.6B. Management framed agentic AI and inferencing as simultaneously lifting both accelerator and high-performance CPU demand.

Server CPU TAM sharply raised on agentic AI

Management raised its server CPU TAM outlook to greater than 35% annual growth, reaching over $120B by 2030 — roughly double the ~18% CAGR / ~$60B view given only months earlier at the November Analyst Day. The revision reflects agentic AI and inferencing driving incremental CPU compute for orchestration, data movement, and head-node roles alongside GPUs.

AMD framed the CPU opportunity in three buckets — general-purpose, AI head nodes, and agentic AI — and argued a broad portfolio (throughput-, power-, cost- and AI-optimized) is required, reaffirming a >50% share target. The CPU-to-GPU ratio is shifting from historical 1:4–1:8 host-node configurations toward ~1:1 or higher.

Instinct MI450/Helios momentum and 2027 visibility

Data Center AI revenue grew a significant double-digit percentage YoY but was down modestly QoQ due to a China revenue step-down. Management said MI450 series is now sampling to lead customers with lead-customer forecasts exceeding initial plans, a broadening base of new customers, and additional multi-gigawatt opportunities, giving confidence in tens of billions of annual Data Center AI revenue in 2027 and exceeding the >80% long-term CAGR target.

Visibility now extends to which data centers GPUs will be installed in. The Helios rack-scale platform (Instinct GPUs + EPYC Venice CPUs) is on track to ramp production shipments in H2, with initial volume in Q3, a significant ramp in Q4, and continued ramp into Q1.

Client, Gaming and Embedded

Client and Gaming revenue was $3.6B, up 23% YoY. Client rose 26% YoY to $2.9B on Ryzen strength, favorable mix and share gains, with commercial Ryzen Pro sell-through up more than 50% YoY across Dell, HP and Lenovo. Gaming rose 11% YoY to $720M on Radeon 9000 demand, partly offset by semi-custom console-cycle decline.

Embedded returned to growth at $873M (up 6% YoY) with double-digit design-win momentum and billions of dollars in new wins as the portfolio expands from FPGA-centric to adaptive x86 and semi-custom. Management is planning for softer H2 consumer PC and gaming demand on higher memory/component costs.

Gross margin drivers and the MI450 mix headwind

Non-GAAP gross margin was 55% in Q1 (up 170 bps YoY) and is guided to ~56% in Q2, aided by favorable product mix and higher data center contribution. Management pointed to multiple H2 tailwinds — server CPU growth, richer Client mix, lower Gaming, and accretive Embedded — to offset gross-margin dilution as MI450 begins ramping in Q3 and ramps significantly in Q4 at a below-corporate-average margin.

On Instinct, the near-term focus is topline growth, with ASP and (more importantly) cost/scale seen as later margin levers. The long-term GM framework remains 55%–58%.

Supply chain, memory inflation and opex investment

Management repeatedly acknowledged a tight supply chain across wafer, back-end capacity, data-center build-out and power, but expressed confidence in supplying — and exceeding — planned growth, with '27 and '28 CPU demand now being planned. Memory prices are rising; AMD says it has secured sufficient memory supply via deep vendor partnerships, is sharing some cost inflation with customers on server ASPs, and expects the larger demand impact of memory inflation to fall on consumer PC and gaming in H2.

Non-GAAP opex rose 42% YoY to $3.1B; management defended running above prior guidance as demand-driven investment, and signaled R&D will grow faster than SG&A going forward after a period of go-to-market build-out.

Summary of the company's earnings call; see how call summaries are prepared. Not investment advice.