Advanced Micro Devices Financial Model
Semiconductors Company Financials Example (Free Excel Download)
Advanced Micro Devices, Inc. (AMD) is a global semiconductor company that designs and develops high-performance computing and graphics solutions.
professionals from Deloitte
Used by professionals from
About this model
This model forecasts Advanced Micro Devices' revenue growth, margin expansion, and free cash flow generation to determine its equity valuation, focusing specifically on the ramp of its Instinct AI accelerators and EPYC server CPUs against competitive pressures.
Advanced Micro Devices, Inc. (AMD) is a global semiconductor company that designs and develops high-performance computing and graphics solutions. The company operates a fabless business model, relying on third-party foundries like TSMC and GlobalFoundries to manufacture its silicon designs. Business segments include Data Center (approximately 48% of revenue), Client and Gaming (approximately 42% of revenue), and Embedded (approximately 10% of revenue). The company generates the majority of its revenue globally, with significant exposure to US hyperscalers and Asian supply chains. AMD holds a strong competitive position as the second-largest player in x86 CPUs behind Intel, the second-largest player in discrete GPUs behind Nvidia, and the market leader in FPGAs. Recent major events include the $49 billion acquisition of Xilinx in 2022, the acquisition of ZT Systems in 2025 to bolster data centre system capabilities, and a segment reporting change in Q1 2025 that combined the previously separate Client and Gaming segments.
The downloadable Advanced Micro Devices financial model includes SEC-sourced historical financials, forecast assumptions, core operating schedules, and valuation outputs in an Excel workbook built for review and scenario analysis.
A turnkey financial model
Live formulas, no hardcoded values
Outputs are driven by live formulas, so the workbook updates from its assumptions instead of relying on hardcoded results.
All assumptions in one tab
Inputs are clearly marked in the Assumptions tab and separated from calculations, making it clear what to change and what to leave intact.
Statements always balancing
For integrated-statement models, the balance sheet, cash flow, and supporting schedules tie through properly.
Distinct schedules for clarity
Debt, working capital, taxes, and cash flow can get messy quickly. We group calculations in clear schedules, not across disconnected tabs.
No hidden macros or external links
There are no unexplained external workbook links or macros to undermine auditability or portability.
Changes flow through the model
Update a key driver and see the impact carry through the forecast, financing, and return outputs. We never use hardcoded numbers in formulas.
Historicals & AssumptionsAdvanced Micro Devices financial modelCompany, Historicals & Assumptions used
Source: SEC EDGAR · values in USD
| Line item | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 |
|---|---|---|---|---|---|
| Revenue | $16.43B | $23.60B | $22.68B | $25.79B | $34.64B |
| Gross profit | $7.93B | $10.60B | $10.46B | $12.72B | $17.15B |
| Operating income | $3.65B | $1.26B | $401.0M | $1.90B | $3.69B |
| Net income | $3.16B | $1.32B | $854.0M | $1.64B | $4.33B |
Forecast assumptions
Defaults used in the downloadable model. Forecast horizon: FY2026–FY2030.
See 8 moreSee less
How to build a detailed financial model for Advanced Micro Devices
A complete walkthrough of every driver, margin, working-capital input, and capital-allocation assumption used in the downloadable model.
Revenue Deep Dive
Data Center
- Segment name: Data Center
- Revenue driver formula: (Server CPU Volume x Server CPU ASP) + (AI Accelerator Volume x Accelerator ASP)
- Historical growth rate: 94% in FY2024, 32% in FY2025.
- Key growth levers and headwinds: Ramping of the Instinct MI300 and MI350 series GPUs for AI workloads, and continued EPYC CPU market share gains against Intel. Headwinds include aggressive pricing from competitors and TSMC supply constraints.
- Pricing dynamics: High average selling prices (ASPs) for AI GPUs, offset by contractual volume discounts for large hyperscaler customers.
- Revenue recognition notes: Recognised upon shipment or delivery to OEM/hyperscaler customers.
- Seasonality: Typically stronger in the second half of the year as hyperscalers deploy new capital expenditure budgets.
Client and Gaming
- Segment name: Client and Gaming
- Revenue driver formula: (PC CPU Volume x PC CPU ASP) + (Discrete GPU Volume x GPU ASP) + (Console SoC Volume x Console ASP)
- Historical growth rate: 51% in FY2025.
- Key growth levers and headwinds: Driven by the Ryzen AI PC upgrade cycle and enterprise desktop refreshes. Headwinds include the aging Sony PlayStation 5 and Microsoft Xbox Series X/S console cycles, which drag down the gaming portion of the segment.
- Pricing dynamics: Highly competitive consumer pricing environment, though AI-enabled Ryzen processors command a premium.
- Revenue recognition notes: Standard point-in-time recognition upon transfer of control to PC OEMs and retail distributors.
- Seasonality: Strongest in Q3 ahead of the holiday shopping season.
Embedded
- Segment name: Embedded
- Revenue driver formula: Embedded Processor Volume x Blended ASP
- Historical growth rate: Declined 33% in FY2024, with a slight decline in FY2025.
- Key growth levers and headwinds: Driven by aerospace, defence, and communications infrastructure. Headwinds include prolonged inventory digestion in the industrial and automotive end markets.
- Pricing dynamics: Long product lifecycles with stable, contractual pricing.
- Revenue recognition notes: Standard point-in-time recognition.
- Seasonality: Generally less seasonal than consumer segments, driven more by specific industrial deployment schedules.
Cost Structure
Variable Costs / COGS
- COGS includes third-party foundry wafer costs (primarily TSMC), packaging and testing, substrate materials, and inventory write-downs.
- Gross margin range: 49% to 51% on a GAAP basis, and 53% to 55% on a Non-GAAP basis over the last two years.
- Key input costs are leading-edge silicon wafers and advanced packaging (CoWoS) capacity.
- COGS scales linearly with volume, but gross margins benefit from a favourable mix shift towards high-margin Data Center products.
Operating Expenses
- R&D: Typically 23% to 25% of revenue. It covers silicon design, tape-out costs, and software stack development (ROCm). It is highly fixed and driven by engineering headcount.
- SG&A: Typically 9% to 11% of revenue. Covers sales, marketing, and corporate overhead. Exhibits strong operating leverage as revenue scales.
- Depreciation & Amortisation: Massive impact on GAAP financials due to the amortisation of acquisition-related intangibles from the Xilinx deal, running at over $2 billion annually.
- Stock-Based Compensation: Significant expense, typically running at 4% to 6% of revenue, used to retain highly competitive engineering talent.
- Restructuring / one-time charges: Infrequent, but acquisition-related integration costs appear during major M&A events.
Margin Profile
- Gross margin: 49% to 51% GAAP.
- Operating margin: Expanding from approximately 7% GAAP in 2024 to the mid-teens in 2025. Non-GAAP operating margins sit in the 25% to 30% range.
- Net margin: Heavily depressed by amortisation on a GAAP basis, but robust on a Non-GAAP cash basis.
- Margin trend: Expanding due to the rapid growth of the high-margin Data Center segment offsetting lower-margin console gaming revenue.
Balance Sheet Structure
- Total assets: Approximately $68 billion to $70 billion.
- Key asset categories: Cash and short-term investments, accounts receivable, inventory, and massive intangible assets.
- Goodwill & intangibles: Represents roughly 65% to 70% of total assets, stemming almost entirely from the $49 billion Xilinx acquisition.
- Working capital profile:
- Days Sales Outstanding (DSO): 35 to 45 days.
- Days Inventory Outstanding (DIO): 90 to 110 days (elevated recently to secure wafers for the AI accelerator ramp).
- Days Payable Outstanding (DPO): 60 to 75 days.
- Net working capital as % of revenue: Typically 10% to 15%.
- Working capital is a use of cash during high-growth phases due to the need to build advanced node inventory months ahead of sales.
- PP&E: Relatively light due to the fabless model, consisting mainly of testing equipment, server racks for internal R&D, and corporate facilities.
- Right-of-use assets: Material but manageable, primarily related to office and lab space leases.
Capital Expenditure & Investment
- Capex as % of revenue: 2% to 3%.
- Maintenance capex vs. growth capex: Roughly 40% maintenance and 60% growth (investing in internal testing capabilities and R&D server infrastructure).
- Major capex programmes: Expanding internal AI compute clusters for software development.
- Capitalised software: Minimal compared to hardware R&D.
- M&A pattern: A mix of transformational deals (Xilinx) and strategic bolt-on acquisitions (Pensando, Silo AI, ZT Systems).
- Typical acquisition multiple paid: High premiums for strategic AI and networking assets.
Debt & Capital Structure
- Total debt: Approximately $2.5 billion to $3.0 billion, resulting in a negative net debt position given the large cash balance.
- Debt/EBITDA ratio: Near zero (well below 1.0x).
- Credit rating: Investment grade (A- / A3).
- Key debt instruments: Senior unsecured notes with staggered maturities.
- Maturity profile: Well-laddered with no near-term liquidity cliffs.
- Interest rate profile: Primarily fixed-rate bonds with a low weighted average cost of debt.
- Covenants: Standard investment-grade covenants, highly flexible.
- Share repurchase programme: Active programme designed primarily to offset dilution from stock-based compensation, running at $1 billion to $2 billion annually.
- Dividend policy: The company does not pay a dividend.
Cash Flow Characteristics
- Operating cash flow conversion: Strong on a Non-GAAP basis, though GAAP OCF/Net Income appears artificially high because GAAP net income is depressed by non-cash amortisation.
- Free cash flow margin: 15% to 25% of revenue.
- Major non-cash items: Amortisation of acquisition-related intangibles, stock-based compensation, and depreciation.
- Working capital cash flow impact: Inventory builds for new product launches act as a significant use of cash.
- Capex intensity: Very low due to the fabless manufacturing model.
- Cash tax rate: Often differs from the GAAP effective tax rate due to the tax treatment of stock-based compensation and acquisition-related IP realignments.
Sheet Structure
- Assumptions: Hardcoded inputs for segment growth, pricing, margins, working capital days, and macroeconomic drivers.
- Revenue Build: Volume and ASP forecasts broken down by the three reporting segments (Data Center, Client and Gaming, Embedded).
- Income Statement: Consolidated P&L projecting revenue down to GAAP Net Income, with a clear reconciliation to Non-GAAP Net Income (adding back amortisation and SBC).
- Balance Sheet: Assets, liabilities, and shareholders' equity, driven by working capital schedules and retained earnings.
- Cash Flow Statement: Indirect method bridging GAAP Net Income to Operating Cash Flow, Investing Cash Flow, and Financing Cash Flow.
- Working Capital Schedule: Calculations for accounts receivable, inventory, and accounts payable based on DSO, DIO, and DPO.
- Depreciation & Amortisation Schedule: Waterfall schedules for PP&E depreciation and the run-off of Xilinx-related intangible assets.
- Debt Schedule: Tracking of senior notes, interest expense calculations, and interest income on cash balances.
- DCF Valuation: Unlevered free cash flow build, WACC calculation, terminal value, and implied share price.
Key Financial Relationships
- Data Center Revenue = (Instinct GPU Shipments x Blended GPU ASP) + (EPYC CPU Shipments x Blended CPU ASP)
- Client and Gaming Revenue = (Ryzen CPU Shipments x Ryzen ASP) + (Discrete GPU Shipments x GPU ASP) + (Semi-Custom Console SoC Shipments x SoC ASP)
- Embedded Revenue = Embedded Unit Volume x Blended Embedded ASP
- Total Revenue = Data Center Revenue + Client and Gaming Revenue + Embedded Revenue
- Total COGS = Total Revenue x (1 - Blended Gross Margin %)
- R&D Expense = Prior Year R&D Expense x (1 + R&D Headcount Growth Rate)
- Non-GAAP Operating Income = GAAP Operating Income + Stock-Based Compensation + Amortisation of Acquisition-Related Intangibles
- Accounts Receivable = (Total Revenue / 365) x DSO
- Inventory = (Total COGS / 365) x DIO
- Accounts Payable = (Total COGS / 365) x DPO
- Free Cash Flow = Operating Cash Flow - Purchases of Property and Equipment
- Net Debt = Total Debt - Cash and Cash Equivalents
Cross-Sheet Dependencies
- The Revenue Build sheet feeds the top line of the Income Statement.
- The Income Statement calculates Net Income, which serves as the starting point for the Cash Flow Statement.
- The Working Capital Schedule calculates the period-over-period change in net working capital, which feeds into the operating activities section of the Cash Flow Statement.
- The Cash Flow Statement calculates the net change in cash, which links to the cash line item on the Balance Sheet.
- The Debt Schedule calculates interest expense, which feeds the Income Statement. To prevent circularity, interest expense must be calculated using the beginning-of-period debt balance rather than the average balance.
- The Depreciation & Amortisation Schedule feeds both the operating expenses on the Income Statement and the non-cash add-backs on the Cash Flow Statement.
Sign Convention
- Revenue, Assets, and Equity are represented as positive numbers.
- Expenses (COGS, R&D, SG&A) and Liabilities are represented as positive numbers within their specific schedules, but are subtracted in aggregation formulas (e.g., Gross Profit = Revenue - COGS).
- On the Cash Flow Statement, cash inflows are positive and cash outflows (such as capital expenditures, debt repayments, and share repurchases) are negative.
Things Most Likely to Go Wrong
- Mismodelling the segment change: AMD combined the Client and Gaming segments in Q1 2025. Historical data prior to 2025 must be aggregated to match the new three-segment structure.
- Forgetting to add back the massive Xilinx amortisation when calculating Non-GAAP metrics. Failing to do this will make the company look artificially unprofitable.
- Underestimating inventory days: The AI GPU ramp requires significant upfront wafer purchases from TSMC, meaning inventory balances will grow faster than historical averages.
- Ignoring stock-based compensation: SBC is a material expense. Excluding it from adjusted figures flatters margins significantly, and it must be accounted for in share count dilution.
- Overestimating capex: Because AMD is a fabless company, its capex should not scale like an integrated device manufacturer (like Intel). Capex should remain a low single-digit percentage of revenue.
- Circularity in the interest schedule: Linking interest income to average cash balances will cause circular reference errors.
- Miscalculating tax rates: GAAP tax rates fluctuate wildly due to acquisition accounting and IP realignments. The model should use a normalised cash tax rate for valuation purposes.
- Double-counting ZT Systems revenue: Ensure that the integration of the 2025 ZT Systems acquisition is properly modelled without artificially inflating organic Data Center growth.
Validation Checks
- Gross margin should remain in the 49% to 55% range based on recent history; flag if the model projects outside this band.
- Capex as a percentage of revenue should not exceed 5% due to the fabless manufacturing model.
- Data Center revenue must account for greater than 45% of total revenue in 2025 and beyond; flag if the mix shifts backward.
- The Balance Sheet must balance perfectly in every projected period: Total Assets = Total Liabilities + Shareholders' Equity.
- Debt/EBITDA should remain below 1.5x, reflecting the company's conservative leverage profile.
- Non-GAAP Net Income should be significantly higher than GAAP Net Income in every period due to the Xilinx amortisation add-back.
- The effective tax rate for Non-GAAP purposes should remain near the management-guided 13% level.
- Share count should slowly increase or remain flat, as share repurchases generally only offset SBC dilution.
Key Assumptions (Default Values)
| Assumption | Default Value | Unit | Rationale |
|---|---|---|---|
| Data Center Revenue Growth | 32.0 | % | FY2025 actual growth rate driven by MI300/MI350 ramp |
| Client and Gaming Revenue Growth | 51.0 | % | FY2025 actual growth rate driven by Ryzen AI PCs |
| Embedded Revenue Growth | -5.0 | % | Continued inventory digestion in industrial end markets |
| GAAP Gross Margin | 51.0 | % | Favourable mix shift toward Data Center products |
| R&D as % of Revenue | 23.0 | % | Historical average, scaling to support AI hardware/software |
| SG&A as % of Revenue | 9.5 | % | Operating leverage on a higher revenue base |
| Days Sales Outstanding (DSO) | 40 | Days | Historical average |
| Days Inventory Outstanding (DIO) | 105 | Days | Elevated to secure advanced packaging for AI GPUs |
| Days Payable Outstanding (DPO) | 65 | Days | Historical average with foundry partners |
| Capex as % of Revenue | 2.5 | % | Standard maintenance for a fabless semiconductor model |
| Effective Tax Rate (Non-GAAP) | 13.0 | % | Management guidance for cash tax rate |
| WACC | 9.5 | % | High beta offset by a low-debt capital structure |
| Terminal Growth Rate | 3.5 | % | Long-term semiconductor industry growth rate |
Data Sources & Benchmarks
- SEC EDGAR: AMD's Investor Relations page for 10-K, 10-Q, and 8-K filings.
- Key peers for benchmarking: Nvidia (NVDA), Intel (INTC), Broadcom (AVGO), and Qualcomm (QCOM).
- Industry data sources: Jon Peddie Research for discrete GPU market share, and Mercury Research for x86 CPU market share.
- Consensus estimates: Bloomberg or FactSet for near-term revenue and EPS consensus to validate model outputs.
Sources
Do more with the Advanced Micro Devices model
Frequently asked
What does Advanced Micro Devices (AMD) do?+
Advanced Micro Devices (AMD) is a global semiconductor company that designs and develops high-performance computing and graphics solutions. The company operates a fabless business model, relying on third-party foundries to manufacture its silicon designs.
What are the primary revenue drivers for Advanced Micro Devices?+
AMD's primary revenue drivers include its Instinct AI accelerators and EPYC server CPUs, which are key components in its Data Center segment. The company also generates significant revenue from its Client and Gaming, and Embedded segments.
What is the assumed annual revenue growth rate in the Advanced Micro Devices financial model?+
The financial model for Advanced Micro Devices assumes an annual revenue growth rate of 20% (0.2). This assumption is crucial for forecasting the company's future financial performance and free cash flow generation over the FY2026–FY2030 horizon.
How does Advanced Micro Devices' fabless model impact its capital expenditure assumptions?+
Due to its fabless business model, AMD's capital expenditure as a percentage of revenue is relatively light, typically assumed to be 2.47%. This reflects investments mainly in internal testing capabilities, R&D server infrastructure, and corporate facilities rather than large manufacturing plants.
What is the main purpose of the Advanced Micro Devices financial model?+
The primary purpose of the Advanced Micro Devices financial model is to forecast the company's revenue growth, margin expansion, and free cash flow generation. This analysis ultimately aims to determine its equity valuation, with a specific focus on the ramp of its AI accelerators and EPYC server CPUs.
Can I download an Excel financial model for Advanced Micro Devices?+
Yes, an Excel financial model for Advanced Micro Devices is available for download. This model forecasts the company's performance from FY2026 through FY2030, providing detailed assumptions for revenue, costs, and capital expenditures to determine its equity valuation.
Have more financial modelling questions? Contact us
Created by ex-finance professionals
Hey, I’m Alex and I created Finamodel.
Over my years in the finance industry I kept building the same models over and over again. Same structure, same assumptions, different logo. So I started building frameworks to turn them into clean, reusable templates.
Every model here is one I’d actually use for a client, and I personally vet each one before it goes up.
I’m not an expert in every industry, but I’ve built enough models to know what belongs in one. And when something is completely foreign to me, I reach out to my network for experts to work on our models with us.
Having a template library on hand cuts a first build from hours to minutes.
Need help finding your model? You’ll find me in the Finamodel app!
Other Semiconductors Company Financial Models
Browse another company in the same sector.
Analog Devices
Analog Devices is a leading global semiconductor company that designs, manufactures, and markets high-performance analog, mixed-signal, and digital signal processing integrated circuits.
Applied Materials
Applied Materials is a global leader in materials engineering solutions, providing the equipment, services, and software used to manufacture semiconductor chips and advanced displays.
Broadcom
Broadcom designs, develops, and supplies a broad range of semiconductor and infrastructure software solutions, combining a fabless semiconductor business with enterprise software.
Intel
Intel Corporation designs and manufactures microprocessors, chipsets, and AI accelerators for personal computers, data centres, and edge devices.
KLA
KLA Corporation is a leading supplier of process control and yield management solutions for the semiconductor and related nanoelectronics industries.
Lam Research
Lam Research Corporation designs, manufactures, and services semiconductor wafer fabrication equipment used in front-end processing.
Microchip Technology
Microchip Technology Incorporated is a broadline supplier of smart, connected, and secure embedded control solutions.
Monolithic Power Systems
Monolithic Power Systems (MPWR) is a fabless global semiconductor company that designs, develops, and markets high-performance, highly integrated power electronics solutions.
Explore more Tech & Software financial model templates.
