Download this white paper to examine the architectural and security considerations associated with specialized networks and learn how data center operators can protect critical data and communications without compromising the performance demanded by distributed AI and HPC workloads.
Rambus Reports Third Quarter 2026 Financial Results
- Record quarterly revenue of $218.3 million, up 22% year over year, exceeding guidance range
- GAAP diluted earnings per share of $0.69 and non-GAAP diluted earnings per share of $0.84, exceeding guidance range
- Record product revenue of $114.3 million, up 15% quarter over quarter and 22% year over year
- Expanded product and IP offerings for next-generation AI systems, including industry-leading PMIC for LPDDR6 SOCAMM2 and data center-class Root of Trust security IP
SAN JOSE, Calif. – October 26, 2026 – Rambus Inc. (NASDAQ:RMBS), a provider of industry-leading chips and IP making data faster and safer, today reported financial results for the third quarter ended September30, 2026. Total revenue for the third quarter was $218.3 million, product revenue was $114.3 million, royalties revenue was $77.0 million and contract and other revenue was $27.0 million. GAAP diluted earnings per share was $0.69 and non-GAAP diluted earnings per share was $0.84. The Company also generated a record $102.5 million of cash from operating activities in the third quarter.
Management Commentary
“Rambus delivered record quarterly revenue and earnings that exceeded our guidance, with strong customer demand driving record product revenue and record cash from operations,” said Luc Seraphin, president and chief executive officer of Rambus. “We continue to expand our product and IP offerings to address the growing performance requirements of next-generation AI and data center infrastructure. Through ongoing execution on our roadmap and accelerating demand for our differentiated solutions, we expect an even stronger performance in Q4 and are well positioned to deliver long-term profitable growth and value for our stockholders.”
| GAAP | Non-GAAP (1) | |||||||||||||||
| Quarterly Financial Review | Three Months Ended September 30, |
Three Months Ended September 30, |
||||||||||||||
| (In millions, except percentages and per share amounts) | 2026 | 2025 | 2026 | 2025 | ||||||||||||
| Revenue | ||||||||||||||||
| Product revenue | $ | 114.3 | $ | 93.3 | $ | 114.3 | $ | 93.3 | ||||||||
| Royalties revenue | 77.0 | 65.1 | 77.0 | 65.1 | ||||||||||||
| Contract and other revenue | 27.0 | 20.1 | 27.0 | 20.1 | ||||||||||||
| Total revenue | 218.3 | 178.5 | 218.3 | 178.5 | ||||||||||||
| Cost of product revenue | 45.5 | 34.3 | 45.3 | 34.2 | ||||||||||||
| Cost of contract and other revenue | 0.8 | 0.5 | 0.8 | 0.5 | ||||||||||||
| Amortization of acquired intangible assets (included in total cost of revenue) | 1.1 | 1.7 | — | — | ||||||||||||
| Total operating expenses | 87.3 | 78.7 | 70.0 | 64.6 | ||||||||||||
| Operating income | $ | 83.6 | $ | 63.3 | $ | 102.2 | $ | 79.2 | ||||||||
| Operating margin | 38% | 35% | 47% | 44% | ||||||||||||
| Net income | $ | 74.6 | $ | 48.4 | $ | 91.7 | $ | 68.2 | ||||||||
| Diluted earnings per share | $ | 0.69 | $ | 0.44 | $ | 0.84 | $ | 0.62 | ||||||||
(1) See “Supplemental Reconciliation of GAAP to Non-GAAP Results” table included below. Note that the applicable non-GAAP measures are presented and that revenue is solely presented on a GAAP basis.
GAAP revenue for the quarter was $218.3 million, which exceeded the Company’s guidance range of $210 million to $216million. The Company had product revenue of $114.3 million, royalties revenue of $77.0 million and contract and other revenue of $27.0 million.
The Company had total GAAP cost of revenue of $47.4 million and operating expenses of $87.3 million. The Company also had total non-GAAP operating expenses of $116.1 million (including non-GAAP cost of revenue of $46.1 million). The Company had GAAP diluted earnings per share of $0.69 and non-GAAP diluted earnings per share of $0.84. The Company’s basic share count was 108 million shares and its diluted share count was 109 million shares.
Cash, cash equivalents, and marketable securities as of September 30, 2026 were $776.9 million, a decrease of $48.0 million as compared to June 30, 2026, mainly due to $135.4 million paid for repurchases of common stock and $11.2 million paid for capital expenditures, partially offset by $102.5 million cash provided by operating activities.
2026 Fourth Quarter Outlook
| (In millions, except per share amounts) | GAAP | Non-GAAP (1) | ||
| Revenue | $227 – $233 | $227 – $233 | ||
| Product revenue | $127 – $133 | $127 – $133 | ||
| Royalties revenue | $68 – $74 | $68 – $74 | ||
| Contract and other revenue | $26 – $32 | $26 – $32 | ||
| Total operating costs and expenses (2) | $143 – $139 | $124 – $120 | ||
| Diluted earnings per share | $0.68 – $0.76 | $0.84 – $0.92 |
(1) See “Reconciliation of GAAP Forward-Looking Estimates to Non-GAAP Forward-Looking Estimates” table included below. Note that the applicable non-GAAP measures are presented and that revenue is solely presented on a GAAP basis.
(2) Total operating costs and expenses include cost of sales and operating expenses.
For the fourth quarter of 2026, the Company expects revenue to be between $227 million and $233 million. The Company also expects product revenue to be between $127 million and $133million, royalties revenue to be between $68 million and $74 million, and contract and other revenue to be between $26 million and $32 million. Revenue is not without risk and achieving revenue in this range will require that the Company sign customer agreements for various product sales and solutions licensing, among other matters.
The Company also expects GAAP operating costs and expenses to be between $143 million and $139 million and non-GAAP operating costs and expenses to be between $124 million and $120 million. The Company expects interest and other income to be $6.5 million. These expectations also assume a GAAP tax rate of 17.5%, a non-GAAP tax rate of 16% and a diluted share count of 109 million shares, and exclude stock-based compensation expense of $18.2million, amortization of acquired intangible assets of $1.0 million and other one-time adjustments of ($0.2) million.
Additionally, the Company expects GAAP diluted earnings per share to be between $0.68 and $0.76 and non-GAAP diluted earnings per share to be between $0.84 and $0.92.
Conference Call
The Company’s management will discuss the results of the quarter during a conference call scheduled for 2:00 p.m. PT today. The call will be audio, slides will be available online at investor.rambus.com, and a replay will be available for the next week at the following numbers: (833) 309-1852 (domestic) or +44 800-007-4044 (international) with ID# 254362271.
Non-GAAP Financial Information
In the commentary set forth above and in the financial statements included in this earnings release, the Company presents the following non-GAAP financial measures:cost of product revenue, operating expenses, operating income, gross margin, operating margin, net income and diluted net income per share (diluted earnings per share). In computing each of these non-GAAP financial measures, the following items were considered as discussed below: stock-based compensation expense, restructuring charges, amortization of acquired intangible assets, facility closure costs, income tax adjustment, and certain other one-time adjustments. The non-GAAP financial measures disclosed by the Company should not be considered a substitute for, or superior to, financial measures calculated in accordance with GAAP, and the financial results calculated in accordance with GAAP and reconciliations from these results should be carefully evaluated. Management believes the non-GAAP financial measures are appropriate for both its own assessment of, and to show investors, how the Company’s performance compares to other periods. The non-GAAP financial measures used by the Company may be calculated differently from, and therefore may not be comparable to, similarly titled measures used by other companies. A reconciliation from GAAP to non-GAAP results is included in the financial statements contained in this release.
The Company’s non-GAAP financial measures reflect adjustments based on the following items:
Stock-based compensation expense. These expenses primarily relate to employee stock purchase plans, and employee non-vested equity stock and non-vested stock units. The Company excludes stock-based compensation expense from its non-GAAP measures primarily because such expenses are non-cash expenses that the Company does not believe are reflective of ongoing operating results. Additionally, given the fact that other companies may grant different amounts and types of equity awards and may use different valuation assumptions, excluding stock-based compensation expense permits more accurate comparisons of the Company’s results with peer companies.
Restructuring charges. These charges include severance, exit costs and other related costs. The Company excludes these charges because such charges are not directly related to ongoing business results and do not reflect expected future operating expenses.
Amortization of acquired intangible assets. The Company incurs expenses for the amortization of intangible assets acquired in acquisitions. The Company excludes these items because these expenses are not reflective of ongoing operating results in the period incurred. These amounts arise from the Company’s prior acquisitions and have no direct correlation to the operation of the Company’s core business.
Facility closure costs.These charges consist of exit costs associated with building leases that were abandoned. The Company excludes these charges because such charges are not directly related to ongoing business results and do not reflect expected future operating expenses.
Income tax adjustment. For purposes of internal forecasting, planning and analyzing future periods that assume net income from operations, the Company estimates a fixed, long-term projected tax rate of approximately 16 percent and 20 percent for 2026 and 2025, respectively, which consists of estimated U.S. federal and state tax rates, and excludes tax rates associated with certain items such as withholding tax, tax credits, deferred tax asset valuation allowance and the release of any deferred tax asset valuation allowance. Accordingly, the Company has applied these tax rates to its non-GAAP financial results for all periods in the relevant years to assist the Company’s planning.
On occasion in the future, there may be other items, such as significant gains or losses from contingencies, that the Company may exclude in deriving its non-GAAP financial measures if it believes that doing so is consistent with the goal of providing useful information to investors and management.
Forward-Looking Statements
This release contains forward-looking statements under the Private Securities Litigation Reform Act of 1995, including those relating to Rambus’ expectations regarding business opportunities, the Company’s ability to deliver long-term, profitable growth, product and investment strategies, and the Company’s outlook and financial guidance for the fourth quarter of 2026 and related drivers, and the Company’s ability to effectively manage market challenges. Such forward-looking statements are based on current expectations, estimates and projections, management’s beliefs and certain assumptions made by the Company’s management. Actual results may differ materially. The Company’s business generally is subject to a number of risks which are described more fully in Rambus’ periodic reports filed with the Securities and Exchange Commission. The Company undertakes no obligation to update forward-looking statements to reflect events or circumstances after the date hereof.
Contact
Sumeet Gagneja
Senior Vice President and Chief Financial Officer
(408) 462-8000
[email protected]
Rambus Inc.
Condensed Consolidated Balance Sheets
(Unaudited)
| (In thousands) | September 30, 2026 |
December 31, 2025 |
||||||
| ASSETS | ||||||||
| Current assets: | ||||||||
| Cash and cash equivalents | $ | 110,531 | $ | 182,826 | ||||
| Marketable securities | 666,376 | 579,005 | ||||||
| Accounts receivable | 142,149 | 137,476 | ||||||
| Unbilled receivables | 26,818 | 25,209 | ||||||
| Inventories | 87,235 | 44,098 | ||||||
| Prepaids and other current assets | 18,323 | 20,202 | ||||||
| Total current assets | 1,051,432 | 988,816 | ||||||
| Intangible assets, net | 5,711 | 10,171 | ||||||
| Goodwill | 286,812 | 286,812 | ||||||
| Property and equipment, net | 107,918 | 113,051 | ||||||
| Operating lease right-of-use assets | 13,702 | 17,112 | ||||||
| Deferred tax assets | 86,500 | 105,542 | ||||||
| Other assets | 6,157 | 8,041 | ||||||
| Total assets | $ | 1,558,232 | $ | 1,529,545 | ||||
| LIABILITIES & STOCKHOLDERS’ EQUITY | ||||||||
| Current liabilities: | ||||||||
| Accounts payable | $ | 23,349 | $ | 35,915 | ||||
| Accrued salaries and benefits | 20,372 | 22,044 | ||||||
| Deferred revenue | 32,391 | 29,980 | ||||||
| EDA tools software licenses liability | 14,795 | 14,884 | ||||||
| Operating lease liabilities | 5,645 | 6,310 | ||||||
| Other current liabilities | 15,543 | 11,441 | ||||||
| Total current liabilities | 112,095 | 120,574 | ||||||
| Long-term operating lease liabilities | 14,572 | 18,671 | ||||||
| Long-term EDA tools software licenses liability | 9,053 | 20,908 | ||||||
| Other long-term liabilities | 4,781 | 4,967 | ||||||
| Total long-term liabilities | 28,406 | 44,546 | ||||||
| Total stockholders’ equity | 1,417,731 | 1,364,425 | ||||||
| Total liabilities and stockholders’ equity | $ | 1,558,232 | $ | 1,529,545 | ||||
Rambus Inc.
Condensed Consolidated Statements of Income
(Unaudited)
| Three Months Ended September 30, |
Nine Months Ended September 30, |
|||||||||||||||
| (In thousands, except per share amounts) | 2026 | 2025 | 2026 | 2025 | ||||||||||||
| Revenue: | ||||||||||||||||
| Product revenue | $ | 114,315 | $ | 93,342 | $ | 301,471 | $ | 250,976 | ||||||||
| Royalties | 76,958 | 65,120 | 230,861 | 207,702 | ||||||||||||
| Contract and other revenue | 27,016 | 20,051 | 73,531 | 58,708 | ||||||||||||
| Total revenue | 218,289 | 178,513 | 605,863 | 517,386 | ||||||||||||
| Cost of revenue: | ||||||||||||||||
| Cost of product revenue | 45,458 | 34,337 | 118,762 | 97,338 | ||||||||||||
| Cost of contract and other revenue | 753 | 531 | 2,659 | 1,708 | ||||||||||||
| Amortization of acquired intangible assets | 1,164 | 1,724 | 4,459 | 5,158 | ||||||||||||
| Total cost of revenue | 47,375 | 36,592 | 125,880 | 104,204 | ||||||||||||
| Gross profit | 170,914 | 141,921 | 479,983 | 413,182 | ||||||||||||
| Operating expenses: | ||||||||||||||||
| Research and development | 49,355 | 49,511 | 150,683 | 138,462 | ||||||||||||
| Sales, general and administrative | 37,684 | 29,155 | 107,616 | 85,328 | ||||||||||||
| Restructuring charges | 223 | — | 3,543 | — | ||||||||||||
| Total operating expenses | 87,262 | 78,666 | 261,842 | 223,790 | ||||||||||||
| Operating income | 83,652 | 63,255 | 218,141 | 189,392 | ||||||||||||
| Interest income and other income (expense), net | 7,162 | 6,327 | 21,373 | 16,411 | ||||||||||||
| Interest expense | (279) | (294) | (860) | (1,053) | ||||||||||||
| Interest and other income (expense), net | 6,883 | 6,033 | 20,513 | 15,358 | ||||||||||||
| Income before income taxes | 90,535 | 69,288 | 238,654 | 204,750 | ||||||||||||
| Provision for income taxes | 15,970 | 20,911 | 36,626 | 38,135 | ||||||||||||
| Net income | $ | 74,565 | $ | 48,377 | $ | 202,028 | $ | 166,615 | ||||||||
| Net income per share: | ||||||||||||||||
| Basic | $ | 0.69 | $ | 0.45 | $ | 1.87 | $ | 1.55 | ||||||||
| Diluted | $ | 0.69 | $ | 0.44 | $ | 1.84 | $ | 1.53 | ||||||||
| Weighted-average shares used in per share calculations: | ||||||||||||||||
| Basic | 107,644 | 107,622 | 108,032 | 107,483 | ||||||||||||
| Diluted | 108,788 | 109,304 | 109,552 | 108,962 | ||||||||||||
Rambus Inc.
Supplemental Reconciliation of GAAP to Non-GAAP Results
(Unaudited)
| Three Months Ended September 30, |
||||||||
| (In thousands, except per share amounts) | 2026 | 2025 | ||||||
| Cost of product revenue | $ | 45,458 | $ | 34,337 | ||||
| Adjustment: | ||||||||
| Stock-based compensation expense | (185) | (180) | ||||||
| Non-GAAP cost of product revenue | $ | 45,273 | $ | 34,157 | ||||
| Total operating expenses | $ | 87,262 | $ | 78,666 | ||||
| Adjustments: | ||||||||
| Stock-based compensation expense | (16,936) | (14,026) | ||||||
| Restructuring charges | (223) | — | ||||||
| Facility closure costs | (78) | — | ||||||
| Non-GAAP total operating expenses | $ | 70,025 | $ | 64,640 | ||||
| Operating income | $ | 83,652 | $ | 63,255 | ||||
| Adjustments: | ||||||||
| Stock-based compensation expense | 17,121 | 14,206 | ||||||
| Amortization of acquired intangible assets | 1,164 | 1,724 | ||||||
| Restructuring charges | 223 | — | ||||||
| Facility closure costs | 78 | — | ||||||
| Non-GAAP total operating income | $ | 102,238 | $ | 79,185 | ||||
| Net income | $ | 74,565 | $ | 48,377 | ||||
| Stock-based compensation expense | 17,121 | 14,206 | ||||||
| Amortization of acquired intangible assets | 1,164 | 1,724 | ||||||
| Restructuring charges | 223 | — | ||||||
| Facility closure costs | 78 | — | ||||||
| Income tax adjustment | (1,489) | 3,867 | ||||||
| Non-GAAP net income | $ | 91,662 | $ | 68,174 | ||||
| Non-GAAP diluted earnings per share | $ | 0.84 | $ | 0.62 | ||||
| Weighted-average shares used in non-GAAP diluted per share calculation | 108,788 | 109,304 | ||||||
Rambus Inc.
Reconciliation of GAAP Forward-Looking Estimates to Non-GAAP Forward-Looking Estimates
(Unaudited)
| 2026 Fourth Quarter Outlook | Three Months Ended December 31, 2026 |
|||||||
| (In millions, except per share amounts) | Low | High | ||||||
| Forward-looking operating costs and expenses | $ | 143.0 | $ | 139.0 | ||||
| Adjustments: | ||||||||
| Stock-based compensation expense | (18.2) | (18.2) | ||||||
| Amortization of acquired intangible assets | (1.0) | (1.0) | ||||||
| Other | 0.2 | 0.2 | ||||||
| Forward-looking Non-GAAP operating costs and expenses | $ | 124.0 | $ | 120.0 | ||||
| Forward-looking net income | $ | 74.7 | $ | 82.9 | ||||
| Adjustments: | ||||||||
| Stock-based compensation expense | 18.2 | 18.2 | ||||||
| Amortization of acquired intangible assets | 1.0 | 1.0 | ||||||
| Other | (0.2) | (0.2) | ||||||
| Income tax adjustment | (1.7) | (1.5) | ||||||
| Forward-looking Non-GAAP net income | $ | 92.0 | $ | 100.4 | ||||
Rambus and ChipAgents Collaborate to Accelerate Chip Design with Agentic AI
Strategic partnership combines Rambus deep semiconductor and silicon IP expertise with ChipAgents tool-agnostic, AI-native chip design platform to improve chip design speed and productivity
- Rambus adopts ChipAgents agentic AI platform to accelerate its own chip and IP design and verification workflows
- The partnership brings together tailored Rambus silicon IP and ChipAgents’ agentic AI capabilities to speed up chip design and IP integration across customer-preferred design environments
- Rambus becomes a strategic investor in ChipAgents, reflecting shared commitment to advancing AI-driven chip design
SAN JOSE and SANTA CLARA, Calif. — Oct. 22, 2026 — Rambus Inc. (NASDAQ: RMBS), a provider of industry-leading chips and silicon IP making data faster and safer, and ChipAgents, a leader in agentic AI platforms for chip design, today announced a multi-faceted strategic agreement to accelerate chip design with agentic AI. As part of the agreement, Rambus is adopting the ChipAgents platform companywide, including deployment across its chip and silicon IP design and verification workflows to accelerate the pace of development. The companies will also collaborate to leverage Rambus’ deep semiconductor expertise and ChipAgents’ agentic AI platform and agentic workflows to enable greater speed and Quality of Results (QoR) in our customers’ chip design process across their preferred design environments. In addition, Rambus has become a strategic investor in ChipAgents, reflecting its confidence in the growing role of AI-native engineering in the future of chip design.
“AI is creating a virtuous cycle of innovation, driving demand for more advanced chips and increasingly complex systems while transforming the way those chips are designed,” said Luc Seraphin, president and chief executive officer of Rambus. “ChipAgents is a fellow industry pioneer and a leader in agentic AI for chip design, making the company a strong strategic investment for Rambus. By adopting the ChipAgents platform, we see opportunities to accelerate the pace of chip and IP development, enhance engineering productivity, and strengthen our long-term position in AI and data-center infrastructure.”
“Rambus has decades of semiconductor chip expertise and a broad portfolio of mission-critical silicon IP for state-of-the-art SoC designs,” said William Wang, founder and chief executive officer of ChipAgents. “Together, we see an opportunity to move beyond automating individual tasks to a more holistic, AI-native approach to chip design. By combining Rambus’ design and IP expertise with our agentic AI platform, we aim to help engineering teams move more efficiently from design intent through silicon implementation while preserving their choice of tools and technologies.”
“Agentic AI can revolutionize complex chip and IP design delivering dramatic benefits in engineering productivity and reduced time to market,” said Simon Blake-Wilson, senior vice president and general manager of Silicon IP at Rambus. “We are excited to adopt the ChipAgents platform across Rambus and apply agentic AI to our chip and silicon IP development workflows. We look forward to working together to drive greater value for our mutual customers by advancing the chip design and IP integration process.”
About Rambus Inc.
Rambus delivers industry-leading chips and silicon IP for the data center and AI infrastructure. With over three decades of advanced semiconductor experience, our products and technologies address the critical bottlenecks between memory and processing to accelerate data-intensive workloads. By enabling greater bandwidth, efficiency and security across next generation computing platforms, we make data faster and safer. For more information, visit rambus.com.
About ChipAgents
ChipAgents is an Agentic AI platform that transforms specifications and code into production-ready RTL, verification assets, and automated root-cause analysis, dramatically shortening design and verification cycles. Founded in 2024, the company brings autonomous AI Agents into production semiconductor workflows, serving both design and verification engineers and embedded systems teams at leading semiconductor companies worldwide. The company is headquartered in Santa Clara, California. For more information, visit chipagents.ai.
Forward-looking statements
Information set forth in this press release, including statements as to Rambus’ outlook and financial estimates and statements as to the expected timing and effects of Rambus products, constitute forward-looking statements within the meaning of the safe harbor provisions of the Private Securities Litigation Reform Act of 1995.
These statements are based on various assumptions and the current expectations of the management of Rambus and may not be accurate because of risks and uncertainties surrounding these assumptions and expectations. Factors listed below, as well as other factors, may cause actual results to differ significantly from these forward-looking statements. There is no guarantee that any of the events anticipated by these forward-looking statements will occur, or what effect they will have on the operations or financial condition of Rambus. Forward-looking statements included herein are made as of the date hereof, and Rambus undertakes no obligation to publicly update or revise any forward-looking statement unless required to do so by federal securities laws.
Major risks, uncertainties and assumptions include, but are not limited to: any statements regarding anticipated operational and financial results; any statements of expectation or belief; other factors described under “Risk Factors” in Rambus’ Annual Report on Form 10-K and Quarterly Reports on Form 10-Q; and any statements of assumptions underlying any of the foregoing. It is not possible to predict or identify all such factors. Consequently, while the list of factors presented here is considered representative, no such list should be considered to be a complete statement of all potential risks and uncertainties.
Press Contacts:
Cori Pasinetti
Rambus Corporate Communications
T: (650) 309-6226
[email protected]
Matt Stubbs
Voxus PR for ChipAgents
T: (801) 703-6626
[email protected]
Rambus Advances Power-Efficient AI Infrastructure with Industry-Leading PMIC Designed for LPDDR6 SOCAMM2
New PMIC expands the Rambus family of LPDDR-based server module solutions and builds on the company’s leadership in advanced server-memory power management
- Designed to enable upcoming low-power, high-performance LPDDR6 SOCAMM2
- Delivers enhanced power efficiency to scale tokens per watt for AI workloads
- Architected in technical cooperation with customers and ecosystem partners to advance future-generation accelerated AI infrastructure
SAN JOSE, Calif. — Oct. 13, 2026 — Rambus Inc. (NASDAQ: RMBS), a premier chip and silicon IP provider making data faster and safer, today announced an industry-leading Power Management IC (PMIC) designed to enable low-power, high-performance LPDDR6 Small Outline Compression Attached Memory Modules (SOCAMM2). Developed in technical cooperation with customers and ecosystem partners to support future-generation accelerated AI infrastructure, the PMIC expands the Rambus family of LPDDR-based server module chips and extends Rambus leadership in advanced server memory products.
As AI training and inference workloads scale, the need for higher throughput within a given power envelope, measured in tokens per watt, is a critical consideration for AI infrastructure. This is driving demand for purpose-built memory solutions optimized for power efficiency, form factor and scalability. SOCAMM2 brings the power and performance benefits of LPDDR technology to a compact, modular and serviceable server-memory architecture designed for demanding AI systems. LPDDR6 SOCAMM2, in development now, represents the next important evolution in LPDDR-based server memory technology.
“Inference and agentic AI are driving exciting new opportunities for memory innovation across the data center,” said Rami Sethi, senior vice president and general manager of Memory Interface Chips at Rambus. “Our new PMIC designed for LPDDR6 SOCAMM2 builds on our leadership in server-memory power management. Working closely with industry leaders in alignment with the broader AI ecosystem, we are helping enable the high-performance, power-efficient memory architectures that maximize tokens per watt for AI workloads.”
“As AI systems place growing demands on memory capacity, bandwidth and power efficiency, LPDRAM is becoming an increasingly important technology for next-generation AI infrastructure,” said Praveen Vaidyanathan, vice president and general manager of Cloud Memory Products at Micron Technology. “Micron has helped pioneer low-power memory for the data center and continues to work closely with ecosystem partners to advance industry standards such as SOCAMM2. We welcome Rambus’ continued investment in LPDDR6 power management solutions, which help strengthen the broader low-power memory ecosystem.”
Building on Rambus Server Memory Power Management Leadership
Rambus provides complete memory interface chipsets for DDR- and LPDDR-based memory modules, including registering clock drivers (RCDs), PMICs, voltage regulators, SPD Hubs and temperature sensors. The company continues to advance its memory module chipset roadmaps for high-volume CPU-based servers while expanding its portfolio to address emerging memory architectures for accelerated AI infrastructure.
Rambus offers a complete LPDDR5X SOCAMM2 chipset, which provides the power delivery, configuration and telemetry functions required by the current-generation SOCAMM2. Building on this chipset strategy, the new PMIC, along with the Rambus SPD Hub, extend the LPDDR-based module roadmap to support the evolving requirements of the upcoming LPDDR6 SOCAMM2.
Optimized for the dynamic workload conditions of future AI systems, the PMIC and SPD Hub designed for LPDDR6 SOCAMM2 provide efficient and precise module-level power management, including advanced power conversion, sequencing, telemetry and fault protection. The PMIC efficiently steps down the system voltage supply to the voltage levels needed to power the LPDDR6 DRAM and all other active chips on the module. These capabilities help support the performance, reliability and power efficiency required for next-generation accelerated computing platforms.
Availability
The Rambus PMIC designed for LPDDR6 SOCAMM2 is currently sampling to lead customers.
More Information:
Learn more about the Rambus SOCAMM2 server module chipset at https://www.rambus.com/memory-interface-chips/socamm2-server-chipset/.
Follow Rambus:
Company website: rambus.com
Rambus blog: rambus.com/blog
LinkedIn: www.linkedin.com/company/rambus
Forward-looking statements
Information set forth in this press release, including statements as to Rambus’ outlook and financial estimates and statements as to the expected timing and effects of Rambus products, constitute forward-looking statements within the meaning of the safe harbor provisions of the Private Securities Litigation Reform Act of 1995.
These statements are based on various assumptions and the current expectations of the management of Rambus and may not be accurate because of risks and uncertainties surrounding these assumptions and expectations. Factors listed below, as well as other factors, may cause actual results to differ significantly from these forward-looking statements. There is no guarantee that any of the events anticipated by these forward-looking statements will occur, or what effect they will have on the operations or financial condition of Rambus. Forward-looking statements included herein are made as of the date hereof, and Rambus undertakes no obligation to publicly update or revise any forward-looking statement unless required to do so by federal securities laws.
Major risks, uncertainties and assumptions include, but are not limited to: any statements regarding anticipated operational and financial results; any statements of expectation or belief; other factors described under “Risk Factors” in Rambus’ Annual Report on Form 10-K and Quarterly Reports on Form 10-Q; and any statements of assumptions underlying any of the foregoing. It is not possible to predict or identify all such factors. Consequently, while the list of factors presented here is considered representative, no such list should be considered to be a complete statement of all potential risks and uncertainties.
Ask the Experts: DDR5 9600 Server RDIMM Chipset
Video: How DDR5 9600 Memory Chipsets Enable Next-Generation Server Performance
As AI, cloud, and enterprise workloads become increasingly memory intensive, server memory must deliver greater bandwidth and capacity while maintaining the reliability data centers demand. In this Ask the Experts video, Rambus memory expert Zaman Mollah explains how DDR5 server memory is evolving to 9600 MT/s and the role memory interface chips play in enabling reliable, high-performance RDIMMs.
Watch the video to learn about the key components of the Rambus DDR5 9600 Server Memory Chipset, including the new sixth-generation Registering Clock Driver (RCD06), and how these technologies support the next generation of AI and data center servers.
What This Video Covers
In this Ask the Experts, Zaman Mollah explores:
- Why AI, cloud, and enterprise workloads are driving the evolution of DDR5 server memory
- What makes operating server memory at 9600 MT/s challenging
- What components make up a DDR5 9600 server memory chipset
- How SPD Hubs, temperature sensors, PMICs, and RCDs work together on an RDIMM
- Why the sixth-generation Rambus RCD06 is critical for 9600 MT/s operation
- How faster DDR5 memory supports increasingly data-intensive AI servers
- How an integrated chipset approach can help simplify server memory design and qualification
What Is Driving DDR5 Server Memory to Higher Speeds?
Modern server processors continue to add more CPU cores and accelerators, while AI, cloud, and enterprise workloads require access to increasingly large amounts of data. That creates demand for both higher memory bandwidth and greater memory capacity.
To keep processors supplied with data, DDR5 server memory must continue increasing performance while meeting the reliability and availability requirements of data center systems. The move toward DDR5-9600, delivering data rates up to 9600 MT/s, is part of that progression.
Why is DDR5 9600 Challenging to Implement?
Increasing DDR5 performance does not simply mean running memory faster. At data rates approaching 9600 MT/s, maintaining signal integrity, power integrity, timing precision, and system reliability becomes increasingly complex.
DDR5 RDIMMs therefore rely on a sophisticated chipset architecture to support high-speed operation while providing the management, monitoring, and control functions required in server environments.
What is a DDR5 9600 Server Memory Chipset?
A DDR5 9600 server memory chipset is a collection of interface and power management components installed on an RDIMM that work together to enable reliable memory operation at data rates up to 9600 MT/s.
The Rambus DDR5 9600 Server Memory Chipset includes four key types of components:
SPD Hub
The Serial Presence Detect (SPD) Hub manages module identification, configuration, and telemetry. It also contains an integrated temperature sensor.
Temperature Sensor ICs
Two discrete Temperature Sensor ICs, combined with the sensor integrated into the SPD Hub, provide three points of thermal telemetry on each RDIMM to support accurate thermal monitoring and management.
PMIC5030
The PMIC5030 power management IC performs efficient voltage regulation directly on the DIMM. Rambus introduced the PMIC5030 for fifth-generation DDR5 memory modules supporting speeds up to 8000 MT/s.
RCD06 Registering Clock Driver
The sixth-generation RCD06 Registering Clock Driver manages command, address, and clock signals between the host processor and DRAM devices, enabling the timing precision and signal quality required for DDR5 9600 operation.
What Does a Registering Clock Driver Do on a DDR5 RDIMM?
The Registering Clock Driver (RCD) is a critical component of the registered DIMM architecture. It receives command, address, and clock signals from the host processor, then buffers, re-drives, and distributes those signals across the RDIMM.
As memory speeds increase, timing margins become more demanding and maintaining signal quality becomes more difficult. The RCD helps the memory subsystem maintain the timing precision needed for reliable high-speed communication between the processor and DRAM devices.
Why is the RCD06 Important for DDR5 9600?
The Rambus RCD06 is the sixth generation of Rambus DDR5 Registering Clock Driver technology and is designed specifically to support the signal integrity and timing requirements of higher-performance DDR5 server memory subsystems.
As the industry progresses toward 9600 MT/s, the RCD plays a central role in maintaining reliable signaling and interoperability. The RCD06 builds on six generations of Rambus DDR5 RCD experience to help module manufacturers and system designers achieve reliable operation in demanding server environments.
How Does DDR5 9600 Support AI Servers?
AI servers are highly data intensive, requiring processors and accelerators to quickly access large amounts of memory. According to the video, the growth of Agentic AI further increases this requirement as processors need fast access to memory across distributed computing architectures.
Higher memory speeds help increase available bandwidth, while the DDR5 chipset provides the telemetry, thermal monitoring, signal management, and power management capabilities needed to operate reliably at scale. Together, these capabilities support both server performance and operational efficiency.
Rambus DDR5 9600 Server Memory Chipset
Rambus provides a complete chipset solution for DDR5-9600 RDIMMs, including the:
- RCD06 Registering Clock Driver
- PMIC5030 Power Management IC
- SPD Hub
- Temperature Sensor ICs
The combination of these components provides the signal management, power delivery, configuration, and thermal monitoring capabilities required for high-performance DDR5 server memory.
Why Choose Rambus for DDR5 Server Memory Chipsets?
Rambus brings decades of high-performance memory interface design experience and works across the server memory ecosystem with processor vendors, memory manufacturers, and system OEMs. Our expertise spans signal integrity, power integrity, interoperability, and reliability.
Rather than supplying only an individual component, Rambus provides a complete chipset solution for every generation of DDR5 RDIMMs. For the 9600 MT/s generation, this includes the SPD Hub, Temperature Sensor ICs, PMIC5030, and RCD06, giving customers an integrated solution from a single supplier.
Watch the Ask the Experts Video
Learn how DDR5 server memory is evolving to 9600 MT/s and how the latest generation of memory interface chips enables greater bandwidth while supporting the reliability, monitoring, and power management requirements of modern servers.
Watch the Ask the Experts video to learn more about the Rambus DDR5 9600 Server Memory Chipset.
Frequently Asked Questions
What is DDR5 9600?
DDR5 9600 refers to DDR5 memory operating at data rates of up to 9600 megatransfers per second (MT/s). Higher DDR5 data rates provide increased memory bandwidth for data-intensive server workloads.
What components are in a DDR5 server memory chipset?
The Rambus DDR5 9600 Server Memory Chipset includes an RCD06 Registering Clock Driver, PMIC5030, SPD Hub, and Temperature Sensor ICs. Together, these components provide signal management, power regulation, configuration, telemetry, and thermal monitoring functions.
What does an RCD do in a DDR5 RDIMM?
A Registering Clock Driver receives command, address, and clock signals from the host processor, buffers and re-drives those signals, and distributes them across the RDIMM with the timing precision required for reliable memory operation.
Why is DDR5-9600 important for AI servers?
AI servers require high memory bandwidth to keep increasingly powerful processors and accelerators supplied with data. DDR5-9600 increases memory performance, while the supporting chipset provides the monitoring, thermal management, power delivery, and signal integrity capabilities needed for reliable operation at scale.
From Caliptra Foundation to Full Production Deployment
Download this white paper to learn how organizations can establish trust across every device in AI infrastructure and modern data centers as heterogeneous processors, accelerators, and controllers become increasingly prevalent. Discover how the open-source Caliptra Root of Trust framework provides a common foundation for device identity, measured boot, and attestation across data center-class systems, and explore the challenges and best practices for transforming an open standard into a production-ready deployment. Learn how enterprises can stay aligned with the Caliptra ecosystem while accelerating implementation, reducing integration risk, and achieving enterprise-grade security for data center and AI infrastructure.
