AMD, founded in 1969, is a fabless designer of CPUs, GPUs, FPGAs, DPUs and rack-scale AI systems, with production outsourced mainly to TSMC.
Latest update Oct 9, 2026
01
Lisa Su has been CEO since 2014. Fiscal 2025 revenue was a record $34.6 billion. Q2 2026 revenue was $11.536 billion, with Data Center $6.718 billion. Helios remains a reference design on AMD's product page, with volume deployments still dated to the second half of 2026. As of 9 October, after Q3 closed, neither AMD nor Oracle had published a customer-available 50,000-GPU Helios supercluster, and OpenAI, Microsoft and Meta had not named a Helios rack receipt. In Seoul on 7 October Lisa Su told reporters AMD had started shipping MI455 and the Helios system; that is a shipment-start claim, not a named customer receipt. The World Labs agreement remains an about $8.2 billion all-stock deal with an unpublished share count based on a 10-day Nasdaq VWAP; named regulators are still unpublished. Taalas.com still reads as an independent company. The White House on 8 October listed AMD among Genesis Mission partners at $500 million in compute credits, beside NVIDIA's $1 billion. Q3 results are scheduled after the close on 3 November. Shares closed 8 October at $620.83, about $1.01 trillion on 1.63 billion shares.
amd.comUpdated Oct 9, 2026
Analysis · 3 min read · Sep 18, 2026The long view on AMD: strategy, technology and what to watch02
Q3-Q4 2026
Lisa Su said at Advancing AI 2026 that Helios is in full production, with shipments starting at the end of Q3 and ramping through Q4. Q3 guidance of about $13 billion is the first earnings print that can show whether rack-scale revenue is real. Watch named customer deliveries versus reference-design status.
Q3 2026
OCI is contracted as the first hyperscaler to offer a public MI450 Helios supercluster, 50,000 GPUs starting in calendar Q3 2026 and expanding in 2027. That is the most specific public GPU count attached to Helios.
Calendar Q3 2026
H2 2026
Meta's first gigawatt is a custom GPU based on MI450, shipped in Helios, with Meta also leading on EPYC Venice. AMD said in Q1 2026 that shipments were on track for the second half. Meta is already testing Helios racks.
H2 2026
The August 6, 2026 definitive agreement to buy inference-chip startup Taalas is still subject to customary closing conditions and regulatory approvals. AMD plans to integrate the dataflow technology with Instinct GPUs.
Oct 2026
Lux at Oak Ridge, on Instinct MI355X, EPYC and Pensando, is expected to be the first operational Genesis Mission platform, with an expected start in October 2026. Discovery, on MI430X and 6th Gen EPYC, follows as the longer-term flagship.
Nov 3, 2026
TargetNovember 3, 2026 after the close, call 5 p.m. ET
Q4 2026
OpenAI has a binding commitment for the first 1 GW of MI450-series GPUs, the first tranche of a 6 GW multi-generation deal. At AAI 2026 it said Helios would come online through deployment partners beginning in Q4 2026, accelerating through 2027. The first warrant tranche vests with that 1 GW delivery.
Nov 2026
On August 4 AMD guided Q3 revenue to about $13 billion, plus or minus $300 million (~41% YoY, ~13% QoQ) and non-GAAP gross margin of about 56%. Consensus around the print (Zacks estimates November 3, 2026) will be the first look at whether Helios is in the numbers.
2026
H1 2027
Anthropic will deploy up to 2 GW of MI450-series GPUs in Helios, with the first gigawatt beginning in H1 2027. AMD's up-to-$5 billion equity investment is tied to that compute partnership. Watch whether the first GW date holds as Anthropic's other Nvidia, Google and Amazon capacity comes online.
2027
AMD's public roadmap puts MI500-series GPUs in 2027, paired with EPYC Verano CPUs and Pensando Como/Monza networking in Helios 500. OpenAI has already said it is looking at MI500. MI600 / Helios 600 is dated 2028.
H2 2027
Europe's next LUMI system, a 387.8 million euro EuroHPC machine in Kajaani, is scheduled for H2 2027 on Instinct MI430X and 256-core 6th Gen EPYC, aiming for 10x the AI capacity of today's LUMI.
03
NVIDIA said on Oct. 8, 2026 it will commit $1 billion over five years to U.S. super-intelligence research, including university compute, quantum work, and cloud providers serving government missions, at the White House Science: A New Golden Age event. The White House listed that pledge inside a $2.4 billion industry pool that also names AMD ($500 million), OpenAI ($200 million), Anthropic and Google ($150 million each), and AWS ($50 million). NVIDIA did not split the $1 billion among those uses, and the figure is a multi-year commitment rather than a booked payment.
nvidianews.nvidia.com www.whitehouse.gov + Nvidia+ OpenAI+ Anthropic+ Alphabet+ Amazon
AMD said on September 28 it signed a definitive all-stock agreement to acquire San Francisco world-model lab World Labs, led by Fei-Fei Li, for about $8.2 billion. After close, Li is to join AMD as executive vice president and chief scientist reporting to CEO Lisa Su, and the World Labs team is to keep working on spatial-intelligence models. The deal is expected to close by the end of 2026, subject to regulatory approvals; World Labs is untracked here.
AMD said on September 24 that Perplexity's Portable Computer local agent software now runs on Windows PCs powered by AMD Ryzen AI Max Series processors, including the Ryzen AI Halo developer platform. Perplexity, an untracked software partner, said local inference does not consume Computer credits and is available to Pro and Max subscribers. Jon Peddie Research reported a 24 GB GPU-accessible memory requirement and that neither company published performance figures for the Ryzen AI Max configuration.
TSMC is on track to reach 180,000 3nm wafers per month by early Q4 2026, two to three months ahead of its prior schedule, on strong demand from customers including Nvidia, AMD and Broadcom, according to an early-August 2026 TrendForce report. The company also expects 2nm capacity to reach 100,000 wafers a month by year end, building on the record $60-64 billion 2026 capital spending guidance it raised in July.
AMD and Anthropic announced on July 22 a strategic partnership under which Anthropic will deploy up to 2 gigawatts of AMD Instinct MI450 Series GPUs, with AMD committing a strategic equity investment of up to $5 billion in the AI developer. Deployment of the first gigawatt is set to begin in the first half of 2027, using AMD Helios rack-scale systems built on Instinct MI455X GPUs, EPYC Venice CPUs and Pensando networking. The agreement, which covers tens of billions of dollars of chips, broadens Anthropic's compute stack to four suppliers alongside Nvidia, Google TPUs and Amazon Trainium as it scales Claude ahead of a possible fall IPO.
04
AMD's first rack-scale AI system: 72 Instinct MI455X GPUs in 18 Open Rack Wide compute trays, 18 6th Gen EPYC Venice CPUs, Pensando Vulcano 800 AI NICs, UALink-over-Ethernet scale-up and Ethernet scale-out, running ROCm. A rack is specified at up to 2.9 exaflops FP4, 1.4 exaflops FP8, 31 TB of HBM4 and 1.7 PB/s of memory bandwidth. AMD claims up to 30% more inference tokens per dollar, 15% more peak FP4, and 50% more HBM than Nvidia's Vera Rubin NVL72.
AMD's data-center GPU line. MI350-series (CDNA 4), led by MI355X (288 GB HBM3E, 8 TB/s, up to 10.1 PFLOPS MXFP4, 1,400 W TBP), is the generation shipping into 2026 and the volume driver of Instinct revenue. MI400-series (CDNA 5), launched July 23, 2026, splits into MI455X for frontier AI and AI factories (432 GB HBM4, 23.3 TB/s, up to 40 PFLOPS FP4, 320 billion transistors) and MI430X for HPC and sovereign AI (up to 288 TFLOPS FP64). ROCm and ROCm.ai are the software stack.
6th Generation EPYC 9006-series server CPUs on Zen 6, launched at Advancing AI 2026 as the host CPU for Helios and as a general-purpose and agentic-AI server line. Top SKU EPYC 9996 is specified at 256 cores / 512 threads. AMD positions them as the densest x86 CPU for agentic workloads and the CPU that keeps Instinct GPUs fed. Meta is a lead customer; Azure is adding Venice-powered HDv2 and HXv2 VM series.
AMD's client CPU and APU brand, from desktop AM5 chips through Ryzen AI notebooks and Ryzen AI Halo agentic PCs. Client revenue was a record $10.6 billion in 2025 and $3.1 billion in Q2 2026. Recent additions include Ryzen AI PRO 400 Series, Ryzen AI Halo developer systems, 3D V-Cache on Ryzen PRO workstations, and a commitment to support Socket AM5 through 2029.
RDNA 4 desktop GPUs launched in 2025 as RX 9070 XT ($599 SEP) and RX 9070 ($549 SEP), later joined by RX 9070 GRE and RX 9060 XT. FY2025 gaming revenue was $3.9 billion, but Q2 2026 gaming fell to $779 million (-31% YoY) on lower semi-custom console SoC sales and higher board prices that Lisa Su said weighed on demand.
05
(FY2025, ended Dec 27, 2025, +34% YoY; GAAP net income $4.335B). H1 2026 added $21.789B. Q3 2026 guided to ~$13B plus or minus $300 million
(Q2 2026, ended Jun 27, 2026, +50% YoY, +13% QoQ); Q3 guided to ~$13B plus or minus $300 million
Up to 6 GW OpenAI + up to 6 GW Meta + up to 2 GW Anthropic, plus Oracle's 50,000-GPU Helios cluster and HUMAIN's path to 1 GW by 2030
Up to 160 million shares each at $0.01, vesting on GPU-purchase and stock-price milestones (final OpenAI tranche tied to $600/share)
Approximately 31,000 in the global workforce as of Dec 27, 2025 (FY2025 10-K; later 10-Qs do not disclose a newer headcount)
share count unknown, 10-day Nasdaq VWAP ending 2nd trading day before close
No named customer-available 50,000-GPU cluster or named volume shipment as of 2 Oct 2026
06
Operating status
Operational; World Labs deal signed, close pending
Q3 2026 closed 30 September without a named Helios customer receipt. AMD's Helios page still calls the rack a reference design, not a product for sale, with volume deployments expected in 2H 2026. Oracle GPU pages opened in this pass returned 404 and do not show a live 50,000-GPU public cluster. Lisa Su told reporters in Seoul on 7 October that AMD had started shipping MI455 and the Helios system; that is not a named OpenAI, Microsoft, Meta or Oracle rack. The World Labs deal remains pending: about $8.2 billion all-stock, close expected by the end of 2026 subject to unnamed regulatory approvals, share count still the 10-day Nasdaq VWAP formula. taalas.com still presents Taalas as independent. The White House on 8 October listed AMD at $500 million of Genesis Mission compute credits. Q3 results are scheduled after the close on 3 November 2026. Shares closed 8 October at $620.83, about $1.01 trillion.
Data Center (Instinct, EPYC, networking, FPGAs)
AI accelerators, server CPUs, DPUs/AI NICs, and adaptive SoCs for hyperscalers, AI clouds, enterprises and HPC. $16.635B in FY2025 (+32%) and $6.718B in Q2 2026 (+107%), now 58% of company revenue.
Desktop and notebook CPUs and APUs, including Ryzen AI and Ryzen AI Halo agentic-PC platforms. Client revenue was $10.640B in FY2025 (+51%) and $3.062B in Q2 2026 (+23%).
Gaming (Radeon and semi-custom SoCs)
Radeon RX discrete GPUs and semi-custom console SoCs (PlayStation, Xbox). Gaming was $3.910B in FY2025 (+51%) but $779 million in Q2 2026, down 31% YoY on lower semi-custom revenue and higher graphics-card prices.
Embedded (Versal, Kria, Ryzen Embedded)
Adaptive SoCs, FPGAs, SOMs and embedded x86 for industrial, automotive, communications, aerospace and defense, and physical AI. Embedded was $3.454B in FY2025 (-3%) and $977 million in Q2 2026 (+19%).
07
How past announced dates turned out.
4/4announced dates hit on time
Put Helios into production in 2026
Target 2026 · Resolved Jul 23, 2026
AMD declared Helios in full production at Advancing AI 2026 on July 23, 2026, and launched the MI400 series the same day. Volume customer shipments were still described as late Q3 into Q4 2026.
Announce a multi-gigawatt OpenAI GPU deal
Target 2025 · Resolved Oct 6, 2025
OpenAI signed a 6 GW multi-generation Instinct deal on October 6, 2025, with a binding first 1 GW of MI450 in H2 2026 and a warrant for up to 160 million AMD shares. Hardware delivery is a separate upcoming milestone.
Close the Xilinx acquisition
Target 2022 · Resolved Feb 14, 2022
The all-stock Xilinx deal closed in February 2022 at an estimated $50 billion, adding FPGAs and adaptive SoCs that now sit in the Embedded segment and in Kria physical-AI products.
Launch Ryzen on Zen and restore x86 competitiveness
Target 2017 · Resolved Mar 2, 2017
First-generation Ryzen shipped in March 2017 and began a multi-year recovery of desktop and later server share against Intel, the foundation of today's EPYC and Ryzen businesses.
08
2026
White House lists AMD at $500 million for Genesis Mission credits
A White House fact sheet dated 8 October 2026 listed AMD at $500 million among $2.4 billion of Genesis Mission Consortium compute credits and tools from eleven firms. NVIDIA's own release the same day was a $1 billion five-year US-science commitment. Neither document is a Helios rack receipt. AMD already has Lux as a Genesis platform expected to start in October 2026.
Su says MI455 and Helios shipments have started
In Seoul on 7 October 2026, before a meeting with Samsung Vice Chair Jun Young-hyun, Lisa Su told reporters AMD had started shipping the next-generation MI455 and the Helios system, both using HBM4. The article does not name Oracle's 50,000-GPU public cluster as generally available and does not name an OpenAI, Microsoft or Meta rack receipt.
Agrees to acquire World Labs for about $8.2 billion
AMD announced a definitive all-stock agreement to acquire World Labs, a San Francisco lab led by Fei-Fei Li that builds spatial-intelligence models for generating, reconstructing and simulating 3D environments and for robotic learning. Li is to join AMD as executive vice president and chief scientist reporting to CEO Lisa Su. The transaction is valued at about $8.2 billion and is expected to close by the end of 2026, subject to regulatory approvals. World Labs is untracked in this catalog and is described only as the acquisition target.
Perplexity Portable Computer comes to Ryzen AI Max PCs
AMD and Perplexity said Portable Computer is available on supported Windows PCs powered by Ryzen AI Max Series processors, including the Ryzen AI Halo developer platform. Users can run local AI tasks and recurring workflows on-device without using Perplexity Computer credits for inference. Perplexity's own post listed Pro and Max subscribers, Windows 10 or 11, at least 24 GB of GPU-accessible memory, and PPLX 27B / Qwen 27B models. This is client software on existing Ryzen AI Max hardware, not a Helios or Instinct shipment.
First $1 trillion market capitalization
On 21 September 2026 AMD closed at $615.52 after a 9.95% gain, taking market value through $1 trillion for the first time and becoming the fourth U.S. chipmaker to do so after Nvidia, Broadcom and Micron. Reuters had the stock up 9.6% at $613.31 during the session; CNBC cited an intraday high of $615.52. The shares closed 23 September at $614.61; stockanalysis still showed a $1.00 trillion cap on about 1.63 billion shares. Helios customer shipments, Oracle's planned 50,000-GPU public cluster, and OpenAI/Meta warrant vesting remain the tests of whether the valuation holds.
Wins Europe's LUMI-AI supercomputer on MI430X
EuroHPC JU signs Bull to deliver LUMI-AI in Kajaani, Finland, in H2 2027 on Instinct MI430X GPUs and 256-core 6th Gen EPYC CPUs. The 387.8 million euro system is expected to deliver 10x the AI capacity of the current LUMI machine.
HUMAIN MI355X cluster goes live in Saudi Arabia
On August 31, 2026, AMD, Cisco and HUMAIN say Instinct MI355X systems with EPYC CPUs and Cisco Silicon One networking are in production in the Kingdom. The JV still targets up to 250 MW of MI400-series infrastructure from 2027 and up to 1 GW by 2030.
Agrees to acquire Taalas for inference silicon
On August 6, 2026, AMD announces a definitive agreement to buy Toronto-based Taalas, a 2023-founded inference-chip startup, to fold specialized inference dataflow hardware into Instinct and Helios. The deal is subject to customary closing conditions.
2026
Q2 record $11.5 billion; Data Center more than doubles
On August 4, 2026, AMD reports Q2 revenue of $11.536 billion (+50% YoY), GAAP EPS $1.38 and non-GAAP EPS $1.66. Data Center is $6.718 billion (+107%) and 58% of revenue. Client is $3.062 billion (+23%); gaming $779 million (-31%). Q3 is guided to about $13 billion, plus or minus $300 million, with non-GAAP gross margin about 56%.
Advancing AI 2026: Helios and MI400 Series launch
On July 23, 2026, AMD puts Helios in full production and launches Instinct MI400-series GPUs (MI455X for frontier AI, MI430X for HPC and sovereign AI), 6th Gen EPYC Venice CPUs, ROCm.ai, Ryzen AI Embedded X100 and the Kria AI robotics platform. Helios is claimed to deliver up to 30% more inference tokens per dollar than Nvidia's Vera Rubin NVL72. OpenAI says it expects to bring Helios online in Q4 2026.
Anthropic deal: up to 2 GW of MI450 and up to $5 billion AMD equity
On July 22, 2026, AMD and Anthropic agree that Anthropic will deploy up to 2 gigawatts of Instinct MI450-series GPUs in Helios racks, with the first gigawatt beginning in H1 2027, building on existing MI355X use. AMD commits to a strategic equity investment of up to $5 billion in Anthropic, and the companies will use Claude to optimize ROCm while AMD adopts Claude internally.
Q1 revenue $10.3 billion as Data Center becomes the growth engine
On May 5, 2026, AMD reports Q1 revenue of $10.253 billion (+38% YoY). Data Center is $5.775 billion (+57%). Lisa Su says customer forecasts for MI450 and Helios exceed initial plans. Q2 is guided to about $11.2 billion.
Meta commits to up to 6 GW, including a custom MI450 GPU
On February 24, 2026, AMD and Meta announce plans to deploy up to 6 gigawatts of Instinct GPUs, with the first 1 GW powered by a custom MI450-based GPU. Meta is also a lead customer for 6th Gen EPYC Venice and Verano. AMD's later 10-Q says it issued Meta a warrant for up to 160 million shares on the same $0.01, milestone-vesting structure as OpenAI.
Record FY2025: $34.6 billion revenue, +34%
On February 3, 2026, AMD reports 2025 revenue of $34.639 billion, GAAP net income of $4.335 billion and non-GAAP EPS of $4.17. Data Center is $16.6 billion (+32%), Client and Gaming $14.6 billion (+51%). Results include about $440 million of MI308 China export-control inventory charges.
2025
Oracle to launch a 50,000-GPU MI450 supercluster
On October 14, 2025, Oracle and AMD say OCI will be the first hyperscaler to offer a public AI supercluster on Helios, with an initial 50,000 Instinct MI450-series GPUs starting in calendar Q3 2026 and expanding in 2027.
OpenAI commits to 6 GW of AMD GPUs
On October 6, 2025, AMD and OpenAI announce a multi-year, multi-generation deal for 6 gigawatts of Instinct GPUs, with the first 1 GW of MI450 series starting in H2 2026. AMD issues OpenAI a warrant for up to 160 million shares at $0.01, vesting on purchase and stock-price milestones up to $600 a share. CFO Jean Hu says the partnership is expected to deliver tens of billions of dollars in revenue.
2024
2022
2017
2014
2009
2006
1982
1972
1969
09
