Our guidance incorporates the order trends that we've seen to date and what we believe today to be appropriate assumptions. Our guidance assumes, among other things, that we don't conclude any additional business acquisitions, restructurings, or legal settlements. It's not possible to accurately predict demand for our goods and services, and therefore, our actual results could differ materially from our guidance. Revenue growth of 36.7% year-over-year, accelerating for the fifth straight quarter, our fastest growth in 18 quarters back when AWS was less than half its current revenue size.
We added over $4.6 billion in revenue quarter-over-quarter, about 80% more than our largest increase ever. AWS is now a $169 billion annualized revenue run rate business, which, for perspective, would place it 24th on the Fortune 500 list if it was a standalone company. Our chips business now has an annual revenue run rate of over $25 billion, growing triple-digit percentages year-over-year. Our AI revenue run rate climbed significantly quarter-over-quarter, and is now also over $25 billion, growing triple-digit percentages year-over-year.
We're seeing strong growth across both AI and non-AI, what we call core, and growth in one is driving growth in the other. Growth in AI drives core because post-training reinforcement learning and agent tool use is mostly done on CPUs versus AI accelerators. Then it takes action: scheduling meetings, drafting and sending email, updating a CRM record, building a dashboard, and more. We also have services like Amazon Connect, our call center service, and AWS Transform, which automates software migration growing quickly.
| Metric | Period | Current guidance |
|---|---|---|
| Net sales | Q3 2026 | $197B-$202B (Prime Day timing shift and ~80 bps FX headwind mask underlying strength; ex-Prime Day growth ~400 bps higher) |
| Operating income | Q3 2026 | $22.5B-$26.5B |
| Cash CapEx | FY2026 | ~$200B and higher, with clear line of sight to strong financial returns; majority for AI and AWS |
| AWS power capacity | By end of 2027 | Reaffirmed on track to double 2025 power capacity by year-end 2027 |
| Robotic-arm fleet | FY2026 | Expect to more than double the fleet of robotic arms (e.g., Cardinal and Sparrow) |
| Metric | YoY | Note |
|---|---|---|
| Worldwide revenue | +20% to $200.6B | Broad-based strength across stores and AWS, including a Prime Day timing shift into Q2 for most large countries; 17% worldwide paid-unit growth and double-digit Prime membership growth. |
| Operating income | +43% to $27.5B | Strong AWS profitability plus ~$1.2B of one-time benefits (tariff refunds and an energy-derivative fair-value gain); disciplined cost-to-serve management. |
| AWS revenue | +36.7% to $42.2B | Fifth straight quarter of acceleration and fastest growth in 18 quarters, driven by both AI and core as enterprises migrate to the cloud and AI pulls along core consumption. |
| AWS operating income | +650 bps margin to $16.6B | Disciplined efficiency gains, capacity optimization, custom silicon and network gear, and fixed-cost management (~520 bps ex the derivative gain). |
| North America segment | +16% to $116.2B revenue | Operating income of $9.1B at a 7.9% margin; included the ~$600M tariff-refund benefit. |
| International segment | +15% to $42.2B revenue (ex-FX) | Operating income of $1.7B at a 4.1% margin on continued cost-to-serve progress. |
| Cash CapEx | $53.1B in the quarter | Primarily AWS and generative AI to support strong demand; management sees strong long-term revenue and free-cash-flow potential. |
| Topic | Previous mention | Current period | Trend |
|---|---|---|---|
| AI investment ROIC | Heavy CapEx questioned for returns | Andy Jassy laid out the economics: data-center capital monetizes for 30+ years once servers plug in; servers/networking break even in under three years on 5-6 year lives with most AI capacity contracted for at least five years, giving clear line of sight to strong returns even at ~$200B+ CapEx. | — |
| AWS demand and capacity | Double 2025 power capacity by end of 2027 | Reaffirmed the doubling target; the lion's share of 2027 capacity is already reserved and a meaningful amount of 2028 capacity is reserved, supported by a $496B backlog growing triple digits. | — |
| Custom silicon (Trainium and Graviton) | Building leading price-performance chips | Chips run rate exceeds $25B; Trainium has multi-year, multi-gigawatt commitments from Anthropic and OpenAI plus more startups/enterprises; Graviton is used by 98% of the top 1,000 EC2 customers, with commitments up nearly 3x sequentially; Amazon is exploring selling Trainium to third-party data centers. | — |
| Agentic AI application layer | Providing model building/inference via Bedrock | Expanding up the stack with Bedrock Agents, Amazon Q (now with autonomous background agents and 16 new integrations), Kiro coding agent (usage tripled QoQ, up to 50% more cost-effective), AWS Transform, Amazon Connect, and the new AWS Continuum security service. | — |
| Own frontier model | — | Jassy said AWS can be wildly successful without its own frontier model since no single model will dominate, but Amazon is pursuing one for cost control, prioritization, and speed; within a few years he expects at least a half-dozen comparably good models, all in Bedrock, one of them Amazon's. | — |
| Everyday essentials and grocery | Experimenting with physical formats | Found a needle-mover in same-day perishables now in 2,300 U.S. cities (nine of top-10 sellers are perishables; +50% monthly active perishable customers YTD); grocery GMS was over $150B last year, making Amazon the second-largest U.S. grocer, with Whole Foods and the new Daily Shop format expanding rapidly. | — |