AI Industry News: AMD Q3 Data Center Revenue Up 97% to $8.5 Billion; Q4 Revenue Seen at About $17 Billion

Two Sheep compiles major developments in the AI industry from company announcements, media reports and other public information. This article covers the third-quarter 2026 (July–September) results that chipmaker AMD (Advanced Micro Devices) reported on November 3, as of November 4, 2026.

Key Points

• AMD's July–September 2026 revenue was $13.4 billion, up 45% year over year. GAAP net income was $2.4 billion, and non-GAAP diluted earnings per share were $2.09
• Data Center segment revenue was a record $8.5 billion, up 97% year over year and about 64% of total revenue. AMD attributes this to demand for EPYC CPUs and the launch of Instinct MI450 Series GPUs
• AMD expects fourth-quarter revenue of about $17 billion (±3%), up about 66% year over year at the midpoint. CEO Lisa Su says growth will accelerate as production of Helios racks and EPYC Venice ramps
• Anthropic, Meta, Microsoft, OpenAI and Oracle's cloud are using the first Helios systems. AMD has agreements with OpenAI for 6 GW, Meta for up to 6 GW and Anthropic for up to 2 GW
• The Data Center segment is about one-tenth the size of Nvidia's Data Center segment in its latest quarter ($89.0 billion; different periods). In Japan, SoftBank is jointly verifying AMD Instinct GPUs

1. Results Overview

On November 3, 2026 (US time), AMD reported results for the three months ended September 26 (the third quarter of 2026).

• Revenue: $13,377 million (about $13.4 billion), up 45% from $9,246 million a year earlier and up 16% from $11,536 million in the previous quarter
• Gross margin: 54% on a GAAP basis and 56% on a non-GAAP basis (up 2 points year over year)
• Operating income: $2,832 million on a GAAP basis (up 123% year over year) and $3,923 million non-GAAP (up 75%)
• Net income: $2,428 million on a GAAP basis (up 95%) and $3,471 million non-GAAP (up 77%)
• Diluted earnings per share: $1.46 on a GAAP basis and $2.09 non-GAAP (up 74%)
• Against guidance: In August, AMD guided July–September revenue to about $13 billion (±$300 million), and the actual figure came in above the top of that range

The gap between GAAP and non-GAAP net income includes a $1,097 million non-operating charge for the change in fair value of warrants issued to certain customers. This charge is driven mainly by AMD's share price and other valuation assumptions. The release does not name the customers, but AMD has announced warrants issued to OpenAI and Meta.

Free cash flow for July–September was $554 million, down from $1,558 million in the previous quarter. Revenue for the first nine months was $35,166 million, already above AMD's full-year 2025 revenue of $34.6 billion.

This article is for information only and does not recommend investing in AMD shares.

2. Revenue by Segment

• Data Center: A record $8,546 million, up 97% from $4,341 million a year earlier and up about 27% from $6,718 million in the previous quarter. Segment operating income was $2,882 million (versus $1,074 million a year earlier). AMD does not report GPU and CPU revenue separately
• Client and Gaming: $3,722 million, down 8% year over year. Within it, Client (Ryzen for PCs and more) was $2,949 million, up 7%, and Gaming was $773 million, down 40% on lower semi-custom and graphics revenue
• Embedded: $1,109 million, up 29% year over year

According to CNBC, the Data Center share of total revenue rose from 47% a year earlier to 64%.

3. Fourth-Quarter Outlook and Management Comments

• Revenue: About $17 billion (±3%). The midpoint is about 66% above a year earlier and about 27% above the third quarter
• Gross margin (non-GAAP): About 55%, slightly below the 56% of July–September
• CEO comments: Lisa Su said fourth-quarter growth will accelerate further as AMD scales production of the Helios rack-scale system and the 6th Gen EPYC Venice processor
• CFO comments: CFO Jean Hu said the 74% rise in non-GAAP earnings per share reflects operating leverage as the Data Center business scales

CNBC reported that Su said on the earnings call that she expects AMD's server revenue to double in 2027. This comment does not appear in the earnings release, and this article has not confirmed it against AMD's official records.

Market Reaction

CNBC reported that AMD shares rose about 8% in extended trading after the results. AMD's results and fourth-quarter outlook beat all of the LSEG estimates cited by CNBC (non-GAAP EPS of $1.95, revenue of $13.09 billion and fourth-quarter revenue of $16.2 billion).

4. Helios and Major Customers

What Helios Is

Helios is a rack-scale AI system that combines Instinct MI450 Series GPUs with 6th Gen EPYC (Venice) CPUs. In May, AMD announced that Venice was ramping production on TSMC's 2nm process in Taiwan and said Helios was on track for multi-gigawatt deployments beginning in the second half of 2026.

The earnings release says Anthropic, Meta, Microsoft, OpenAI and Oracle Cloud Infrastructure (OCI) are running workloads on the first Helios systems. It also notes HPE's announcement that Vultr will deploy the first Helios by HPE systems, and the EuroHPC Joint Undertaking's selection of MI430X GPUs for the LUMI-AI supercomputer.

Major Agreements Announced by AMD

Gigawatts measure the power capacity of the equipment to be deployed, not orders or revenue. All are multi-year plans; the Meta and Anthropic figures are stated as maximums.

• OpenAI (October 6, 2025): A 6 GW agreement covering multiple generations of Instinct GPUs. The first 1 GW (MI450) begins deployment in the second half of 2026. AMD issued OpenAI a warrant for up to 160 million AMD shares
• Oracle (October 14, 2025): A plan to offer a publicly available AI supercluster with an initial 50,000 MI450 GPUs starting in July–September 2026, expanding in 2027 and beyond (the latest earnings release does not say the initial 50,000-GPU supercluster is publicly available yet)
• Meta (February 24, 2026): An agreement for up to 6 GW, with shipments of the first 1 GW beginning in the second half of 2026, also with a warrant for up to 160 million shares. On October 27, AMD said the first Helios systems were installed and being tested at Meta and announced that AMD will invest $1 billion in Meta's equity offering
• Microsoft (July 20, 2026): Microsoft will deploy Helios (MI455X and Venice) on Azure. No size was disclosed
• Anthropic (July 22, 2026): Deployment of up to 2 GW of MI450 (Helios), with the first 1 GW beginning in the first half of 2027. AMD said it will invest up to $5 billion in Anthropic

Comparing Scale

In its latest quarter (fiscal Q2 2027, ended July 26, 2026), Nvidia's Data Center revenue was $89.0 billion, up 117% year over year. AMD's $8.5 billion in Data Center revenue is about one-tenth of that. The periods differ, however, and the two segments cover different product mixes, so the comparison is only indicative. As of November 4, Nvidia had not yet reported its next quarter.

Developments in Japan

• SoftBank (February 16, 2026): SoftBank Corp. announced that it had begun joint verification of AMD Instinct GPUs as compute for next-generation AI infrastructure, including a mechanism that partitions GPUs across multiple AI applications. This is a verification stage, not an announcement of production deployment
• Fujitsu (November 1, 2024): Fujitsu and AMD signed a memorandum of understanding to form a strategic partnership, targeting joint development by 2027 of AI and HPC platforms that combine Fujitsu's Arm-based FUJITSU-MONAKA CPU with AMD Instinct accelerators

The latest earnings release makes no mention of Japan, and we have not found any announcement of a Helios deployment in Japan.

5. What It Means for Data Centers and Compute

• AMD's AI business is shifting from individual servers to rack-level systems
• In AI compute, AMD as well as Nvidia now has multi-gigawatt plans, increasing the range of equipment data centers must handle
• Agreements stated in gigawatts are multi-year plans, and actual deployment will proceed in stages

We believe that as more data centers bring in dense rack-scale systems from multiple vendors, vendor-neutral on-site work will matter more: moving in and installing heavy racks, cabling high-speed networks, installing optical modules and replacing parts. We will continue to watch deployments in Japan.

6. About Two Sheep Co., Ltd.

Two Sheep Co., Ltd., headquartered in the Telecom Center Building in Aomi, Koto-ku, Tokyo, provides on-site work at data centers and the shipping and installation of IT equipment.

• 24/7/365 Remote Hands: at data centers in Tokyo and Osaka (Equinix TY and OS sites, AT TOKYO CC1 and CC2, NTT and others), we handle server rack-mounting, cabling, part replacement (RMA), status checks and more
• International shipping and import/export customs for IT equipment: we arrange transport and customs clearance from overseas all the way into the data center
• Migrations, connectivity and optical modules: we handle data center migrations, arrange IP transit and private lines, and sell optical modules (ModuleSheep)
• Support in Japanese, English and Chinese: we coordinate between overseas customers and Japanese data centers

Main Sources

• AMD: Third-quarter 2026 financial results (November 3, 2026)
• AMD: Second-quarter 2026 financial results (August 4, 2026)
• AMD: Fourth-quarter and full-year 2025 financial results (February 3, 2026)
• CNBC: AMD third-quarter earnings report (November 3, 2026)
• AMD and OpenAI: Partnership to deploy 6 GW of GPUs (October 6, 2025)
• Oracle and AMD: Expanded partnership (October 14, 2025)
• AMD and Meta: Partnership to deploy 6 GW of GPUs (February 24, 2026)
• AMD and Meta: Partnership update (October 27, 2026)
• AMD and Microsoft: Deployment of next-generation Instinct and EPYC (July 20, 2026)
• AMD and Anthropic: Partnership to deploy up to 2 GW of GPUs (July 22, 2026)
• AMD: Production ramp of EPYC Venice on TSMC 2nm (May 21, 2026)
• AMD: Taiwan ecosystem investments of more than $10 billion (May 21, 2026)
• Nvidia: Fiscal Q2 2027 results (August 26, 2026)
• SoftBank Corp.: Start of joint verification with AMD (February 16, 2026, Japanese)
• AMD and Fujitsu: Start of strategic partnership (November 1, 2024)

About This Article

• This article was compiled independently by Two Sheep from company announcements, filings with the SEC, media reports and other public information available as of November 4, 2026. Companies' plans and figures may change.
• This article is for information only and does not recommend any company, security or investment.
• Companies are named only to illustrate industry trends; their mention does not indicate any business relationship with Two Sheep. Names of companies, products and services are trademarks or registered trademarks of their respective owners.
• Contact: Japan@2sheep.jp, or use the contact form on this website.