AMD designs processors for PCs, servers and game consoles, and is now the main alternative to Nvidia in AI accelerators. Data center sales more than doubled last quarter and its Helios AI racks start shipping this half. The shares are up more than 140% this year.
Today’s close against the report’s numbers, updated every weekday. The analysis is as of October 21, 2026; the rating as of October 22, 2026. Both change only when the report is revised.
The scenario model’s bear-to-bull range, the probability-weighted value, the price, its 52-week range and the analysts’ average target. The copper line marks the base case; the fine dotted line is today’s close.
Written on October 21, 2026, ahead of its third-quarter results on November 3. Figures are from the company’s filings and releases unless noted. Not investment advice. The author holds no AMD shares directly; AMD is held inside the VanEck Semiconductor ETF (SMH) in the author’s portfolio.
AMD has four businesses. Data Center (58% of revenue) sells EPYC server processors, where it has taken share from Intel for years, and Instinct AI accelerators. Client sells Ryzen PC chips, Gaming the chips inside the PlayStation and Xbox, and Embedded the Xilinx programmable chips it bought in 2022.
The investment case is AI. With the MI450 series and the Helios rack, AMD sells a whole AI system rather than single chips. OpenAI agreed in October 2025 to deploy 6 gigawatts of AMD GPUs and received a warrant for up to 160 million AMD shares that vests as deployments and the share price hit milestones. In 2026 Anthropic agreed to deploy up to 2 gigawatts of MI450 GPUs in Helios racks, and Meta, Microsoft and Oracle are customers too.
Second-quarter revenue rose 50% and Data Center more than doubled. Gaming fell 31% and Embedded grew 19%. GAAP operating income was $2.0 billion and free cash flow $1.56 billion, 14% of revenue. The shares fell 7% the next day because investors wanted more.
For the third quarter AMD guided to about $13.0 billion of revenue and a 56% gross margin. Lisa Su described Q3 as the start of the Helios ramp, Q4 as a step up and Q1 2027 as a further step up, and management expects Data Center revenue to more than double again in 2027. The stock closed at a record $649.42 on October 6 and has fallen about 15% since. Results come on November 3.
| FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | TTM Jun 2026 | |
|---|---|---|---|---|---|---|
| Revenue | $16.4B | $23.6B | $22.7B | $25.8B | $34.6B | $41.3B |
| Operating margin | 22.2% | 5.4% | 1.8% | 7.4% | 10.7% | 15.7% |
| Net income | $3.2B | $1.3B | $0.9B | $1.6B | $4.3B | $6.4B |
| Operating cash flow | $3.5B | $3.6B | $1.7B | $3.0B | $7.7B | $10.1B |
| Capital spending | $0.3B | $0.5B | $0.5B | $0.6B | $1.0B | $1.7B |
| Free cash flow | $3.2B | $3.1B | $1.1B | $2.4B | $6.7B | $8.4B |
GAAP profit looks small next to adjusted figures because AMD still amortises the intangible assets from the Xilinx deal and pays a lot in stock (about $1.9 billion over the past twelve months). The adjusted 2025 figures add both back. This report’s margins are GAAP, after stock pay. The quarterly view shows each quarter.
At the end of June AMD had $13.1 billion of cash and short-term investments and $3.2 billion of debt. The OpenAI warrant is the unusual item: up to 160 million shares, about 10% of today’s count, for a nominal price, with the last tranche tied to a $600 share price. This model counts all of it, so shares rise from 1.63 billion to about 1.80 billion by 2030.
2030 revenue and operating margin, taxed at 21%, times an exit multiple, discounted back four years at 12% (beta 1.6 gives a CAPM cost of about 12.3%), plus net cash and the free cash flow earned along the way (a straight line from about $8.4 billion over the last twelve months to 80% of 2030 after-tax profit). AMD’s own long-term targets, set in November 2025, call for revenue growth above 35% a year and earnings above $20 a share; it now says it expects to beat that figure significantly.
| Bear | Base | Bull | |
|---|---|---|---|
| Revenue in 2030 | $100B | $150B | $200B |
| Operating margin after stock pay | 22% | 30% | 35% |
| Exit multiple | 20× | 25× | 30× |
| Value per share | $148 | $353 | $641 |
| Probability | 25% | 50% | 25% |
Weighted 25/50/25: $374. The base case has revenue about three times this year’s roughly $50 billion: Data Center doubling in 2027, then growth slowing to about 15% by 2030.
Every figure in this valuation comes from one model script, rerun and checked against this page each time the site is built. The XLSX model rebuilds it in live formulas: change any blue input and every year’s cash flow, the bridge to the value, the scenarios and the sensitivity table recalculate.
| To justify the price, you need | Against |
|---|---|
| About $172 billion of revenue in 2030 at a 35% margin and 30× earnings | Close to our bull case |
| Or about $237 billion of revenue at the base margin and multiple | Almost five times this year’s revenue |
The analysts’ average target is $636 and they put the stock at about 34 times their 2027 earnings estimate. Our lower number comes mostly from valuing it on 2030 earnings, after the AI build-out has slowed, rather than on 2027.
| Risk | Why it matters |
|---|---|
| Helios timing | Growth after Q3 depends on racks shipping on schedule; a slip would hit 2027. |
| Nvidia | Nvidia still has most of the market, a software lead (CUDA) and a yearly product cadence. |
| Customer concentration | A few AI labs and cloud providers account for most of the GPU growth. |
| AI spending cycle | If AI data center spending slows after 2027, the highest-growth part of AMD slows with it. |
| Dilution and export rules | The OpenAI warrant adds up to 10% more shares; China export limits already cost $800 million in 2025. |
Upside: a faster Helios ramp, share gains in AI inference, more lab deals like Anthropic’s, and continued EPYC gains in servers.
AMD leads in server CPUs and is now a credible second supplier of AI systems, with signed demand from OpenAI and Anthropic. The price sits between the base and bull cases.
What would change the value: Third-quarter results on November 3 and the size of the fourth-quarter Helios step-up are the next test.
Ranges, not three cases. The expected value is the average of 4,000 runs of this report’s own model, with every input drawn from a range: the base case is the middle, and the bear and bull cases are the 10th and 90th percentiles. Business inputs move together; the discount rate moves on its own. The 25/50/25 weighting in the valuation section gives $374. How the ranges work
Two different questions. The rating is relative: it asks whether the stock is cheaper or dearer than the S&P 500 on the same assumptions. The absolute view asks whether it is worth its price at all: on its own the stock trades above our estimated value, which is 34% below the price.
On the same assumptions every report uses (a 5.15% Treasury yield and a 4.5% equity premium), the S&P 500 itself is worth about 68% of its price. The rating compares each stock with that: Buy if it is at least 15% cheaper than the market on this basis, Sell if at least 15% dearer, Hold in between. It is the same rule for every company, whoever owns it. The rating is the same whether the index is assumed to grow 6% or 10% a year. How ratings work
Against the Street: Analysts average about $636 and value AMD on about 34 times their 2027 earnings. This model values it on 2030 earnings, after the AI build-out has slowed, with a 25× exit multiple in the base case.
| Measure | Advanced Micro Devices | Industry median | Rank | |
|---|---|---|---|---|
| Revenue growth, last fiscal year | 34.3% | 15.3% | 4 of 15 | Better than the median |
| Operating margin | 15.7% | 30.8% | 11 of 15 | Worse than the median |
| Price to earnings | 143.7× | 43.0× | 14 of 14 | Worse than the median |
| Free cash flow yield | 0.9% | 2.5% | 13 of 15 | Worse than the median |
Peers: the 14 other S&P 500 companies in Semiconductors with complete data (ADI, AVGO, FSLR, INTC, MCHP, MPWR, MRVL, MU, NVDA, NXPI, ON, QCOM, SWKS, TXN). Last twelve months where reported, from SEC filings; prices from the latest close. A lower price-to-earnings ratio ranks higher. These figures update with each data run; the rating does not.
The products and customers are there; the shares have run ahead of the revenue.
| What | Figures | Source |
|---|---|---|
| Latest results | $11.5B (Revenue, Q2 2026, +50%), $6.7B (Data Center, +107%), $3.1B (Client, +23%), 56% (Gross margin, adjusted) and others | Q2 2026 results; Ramp and expectations; Analyst views |
| The numbers (five-year table) | Five years of annual figures and the last twelve months | SEC XBRL company facts; latest: 10-K for the period to 2025-12-27, 10-Q for the period to 2026-06-27 |
| Price at writing | $552.72 (Close, Oct 21, 2026) | Yahoo Finance; the time of the quote is stated beside the price |
| Discount rate inputs | 10-year Treasury 5.15%, equity risk premium 4.5% | Fixed for every report so they can be compared; How ratings work |
| Valuation | Every value, case and sensitivity | This site’s model: XLSX, CSV |
| Checked by hand | Not yet | Reports are added to the audit log as they are revised |
| Input | Bear (10th percentile) | Base (median) | Bull (90th percentile) |
|---|---|---|---|
| Revenue, target year | $100B | $150B | $200B |
| Margin, target year | 22.0% | 30.0% | 35.0% |
| Exit multiple | 20× | 25× | 30× |
| Simulation result | |
|---|---|
| Expected value (average of 4,000 runs) | $366.25 |
| Likely range, 10th to 90th percentile | $168.63 to $590.52 |
| Median | $346.07 |
| Chance the value is above the price | 13% |
| Chance it is cheaper than the S&P 500 | 43% |
Each input is drawn from a split normal: its bear and bull values are the 10th and 90th percentiles, so the two sides can differ. Business inputs share a common factor (correlation 0.6); the discount rate is the same in every case, so it is not drawn. Inputs not listed are held at their base value. How the simulation works.