GlobalFoundries and RAAAM Memory Technologies announce the joint development of next-generation GCRAM technology and test chip on FDX platform

If projections hold, GCRAM could cut memory area ~40% and memory power up to 60%, with design access planned for lead customers in early 2027.

Rhea-AI Summary

GlobalFoundries (GFS) and RAAAM Memory Technologies announced a strategic collaboration to develop Gain-Cell RAM (GCRAM) on GF’s FDX FD-SOI platform, achieving a projected 40% memory area shrink compared to commodity SRAM.

The partnership has reached a key milestone with the successful tape-out of a GCRAM test vehicle and a jointly designed test chip, targeting advanced edge AI, automotive and IoT applications. The companies report up to 60% memory power reduction versus SRAM and plan to offer GCRAM design access to lead customers in early 2027. NXP Semiconductors is already evaluating the solution, citing its potential to increase on-chip memory capacity and reduce off-chip data movement for greater system-level efficiency.

Argus Sep 2 session 2 alerts

+0.09% close to close 0.9x rel. volume Open Argus

Details

News Market Reaction – GFS

$24.17B Mkt Cap at Peak Alert

On Sep 2, the day this news came out, GFS closed 0.09% above the previous close. Our momentum scanner recorded 2 alerts for this stock that day.

Data tracked by StockTitan Argus for the Sep 2 session.

Key Figures

Memory area shrink: 40% Memory power reduction: Up to 60% Customer design access: Early 2027

Memory area shrink

40%

Compared to commodity SRAM

Memory power reduction

Up to 60%

Compared to commodity SRAM

Customer design access

Early 2027

Planned next step for lead customers

Historical Context

5 past events · Latest: Aug 10

5 events

1. Aug 10

   AI data-platform partnership

   24h Move

   -7.2%

   Selected Redpanda to unify manufacturing data and develop AI agents.

2. Aug 07

   Partner ecosystem update

   24h Move

   +7.1%

   SEALSQ expanded its quantum initiative ecosystem, including GlobalFoundries.

3. Aug 05

   2Q26 earnings report

   24h Move

   -4.9%

   Reported higher revenue, margins, cash, and third-quarter guidance.

4. Jul 29

   CHIPS award agreement

   24h Move

   -4.0%

   Signed a letter of intent for an expected $300 million R&D award.

5. Jul 01

   Earnings call scheduling

   24h Move

   -6.3%

   Scheduled a conference call to review second-quarter financial results.

24h Move is the share-price change in the day after each event; other market factors may also have contributed.

Key Terms

gain-cell ram, fd-soi, tape-out, bitcell, +1 more

5 terms

gain-cell ram technical

"develop and qualify Gain-Cell RAM (GCRAM) technology"

Gain-cell RAM is a type of dynamic memory that stores bits using a small charge on a capacitor plus a local amplifier or “gain” element that refreshes or senses the charge instead of relying only on large capacitors. Think of it like a tiny inline magnifying glass that keeps a faint mark readable; this lets designers pack more memory into less silicon with lower power or faster access than some older DRAM designs. Investors watch it because the choice of memory architecture affects chip performance, power use, manufacturing cost and competitiveness in devices.

fd-soi technical

"on GF’s industry-leading FDX™ FD-SOI platform"

FD-SOI (Fully Depleted Silicon On Insulator) is a semiconductor manufacturing technique that builds transistor switches on a very thin layer of silicon separated from the bulk wafer by an insulating layer, improving control over electricity flow. For investors, it matters because FD-SOI can deliver faster chips, lower power use and potentially lower production costs—like using a more efficient engine design—affecting a maker’s performance, market competitiveness and profit margins.

tape-out technical

"successful tape-out of a GCRAM test vehicle"

Tape-out is the milestone when a chip designer delivers the finalized, production-ready blueprint of a semiconductor chip to a foundry for fabrication. Think of it as handing a finished blueprint to a factory: it means design work is complete and manufacturing (with its costs and timelines) can begin. For investors, tape-out signals a clear step toward production, potential revenue, and the transition of technical risk into manufacturing and market risk.

bitcell technical

"co-designed the GCRAM bitcell using “pushed” design rules"

A bitcell is the basic memory cell inside a semiconductor chip that stores one binary digit (0 or 1), implemented with transistors and capacitors in DRAM or with small transistor circuits in SRAM. Investors care because a bitcell’s physical size, speed, reliability and manufacturing yield determine a memory chip’s capacity, performance and production cost—key factors for a semiconductor maker’s product competitiveness and margins.

power-performance-area (ppa) technical

"redefines the power-performance-area (PPA) equation"

A power-performance-area (PPA) metric is a shorthand engineers use to describe the three linked characteristics of a semiconductor design: how much electrical power it uses, how fast or capable it is (performance), and how much physical chip area it occupies. Investors track PPA because those three trade-offs affect product competitiveness, manufacturing cost and yield, and ultimately the potential margin and market success of chips—think of it like comparing a car’s fuel use, speed, and size when judging value.

AI-generated analysis. How Rhea-AI works. Not financial advice.

See more from StockTitan in Google Search and AI answers. Adds StockTitan as a preferred source · opens Google

Add on Google Not now

Collaboration targets to enable the densest standard CMOS on-chip memory solution for the FDX platform widely used for advanced AI chips, by delivering 40% area shrink and up to 60% power reduction compared to commodity SRAM

MALTA, N.Y. and PETA TIKVA, Israel, Sept. 02, 2026 (GLOBE NEWSWIRE) -- GlobalFoundries (GF) (Nasdaq: GFS) and RAAAM Memory Technologies, a pioneer in advanced on-chip memory solutions, today announced a strategic collaboration to develop and qualify Gain-Cell RAM (GCRAM) technology. The partnership has reached a major milestone with the successful tape-out of a GCRAM test vehicle on GF’s industry-leading FDX™ FD-SOI platform.

As edge AI, automotive, and IoT devices continue to demand unprecedented amounts of data, on-chip memory has become a critical bottleneck and often constitutes the majority of the silicon area. Designers are looking for innovative on-chip memory solutions offering higher-density and lower-power consumption compared to commodity SRAM.

Leveraging RAAAM’s patented GCRAM, the companies have co-designed the GCRAM bitcell using “pushed” design rules on FDX technology, achieving 40% memory area shrink and up to 60% memory power reduction compared to SRAM. The companies have jointly designed and taped out a GCRAM test chip. At the next step, the companies will start providing GCRAM design access to lead customers in early 2027.

“The successful tape-out of our GCRAM test vehicle on GF’s FDX platform is a testament to the powerful synergy between RAAAM's memory innovations and GlobalFoundries' manufacturing excellence,” said Robert Giterman, CEO and co-founder of RAAAM Memory Technologies. “Our GCRAM technology allows semiconductor companies to significantly increase their on-chip memory capacity or reduce the silicon area while keeping the same amount of on-chip memory, while the ultra-low leakage of FD-SOI technology extends data retention, enabling designers to unlock substantial energy savings for edge AI devices.”

“At GlobalFoundries, we are continuously pushing the boundaries of what is possible on our FDX platform and keep enhancing our IP offering to meet our customers’ needs,” said Sudipto Bose, vice president of FD-SOI product management at GlobalFoundries. “Partnering with RAAAM Memory Technologies allows us to offer our customers an incredibly compelling on-chip memory solution that redefines the power-performance-area (PPA) equation for advanced AI and other applications.”

The semiconductor ecosystem is already recognizing the potential of this joint development. NXP Semiconductors, a leader in intelligent systems at the edge, is closely evaluating the benefits of the GCRAM solution.

“The industry needs a paradigm shift in on-chip memory. The GCRAM solution developed by RAAAM and GF offers immense potential,” said Victor Wang, VP of front end innovation at NXP Semiconductors. “The projected area shrink and power reductions provide a clear path to increasing on-chip memory capacities, which is vital for reducing off-chip data movement and driving greater system-level efficiency.”

About GF

GlobalFoundries (GF) is a leading manufacturer of essential semiconductors, enabling AI at scale from the cloud to the physical world. Through deep partnerships with customers, GF delivers differentiated, power-efficient and high-performance solutions for automotive, aerospace and defense, data center, smart mobile devices, internet of things and other high-growth markets. With global manufacturing operations across the U.S., Europe and Asia, GF is a trusted and holistic technology partner for customers around the world. GF’s talented, global team remains focused every day on security, longevity and sustainability.

For more information, visit www.gf.com.

About RAAAM Technologies

RAAAM Memory Technologies is a fabless semiconductor IP company providing the most cost-effective on-chip memory technology in the industry. RAAAM’s patented Gain-Cell RAM (GCRAM) technology provides up to 50% silicon area reduction and up to 3X reduced power consumption over traditional SRAM. GCRAM can serve as a drop-in SRAM replacement in SoCs and is fully compatible with standard CMOS fabrication flows, allowing designers to integrate more memory in a smaller die area with lower power consumption.

For more information, visit www.raaam-tech.com.

Media contact:
Stephanie Gonzalez
stephanie.gonzalez@gf.com

FAQ

AI-generated questions and answers. How Rhea-AI works. Not financial advice.

What did GlobalFoundries (GFS) and RAAAM Memory Technologies announce about GCRAM technology?

GlobalFoundries and RAAAM Memory Technologies announced a strategic collaboration to develop and qualify Gain-Cell RAM (GCRAM) on GF’s FDX FD-SOI platform and reported a major milestone: successful tape-out of a GCRAM test vehicle and a jointly designed test chip for advanced on-chip memory applications.

How much area and power reduction does the new GCRAM on the FDX platform target for GFS?

The co-designed GCRAM bitcell on the FDX platform targets a 40% memory area shrink and up to 60% memory power reduction compared to commodity SRAM. These gains are aimed at enabling denser, lower-power on-chip memory for edge AI, automotive and IoT devices.

When will lead customers get design access to the GCRAM solution from GlobalFoundries (GFS) and RAAAM?

Lead customers are expected to receive GCRAM design access in early 2027. The companies plan to make the new on-chip memory solution available on the FDX FD-SOI platform after completing current development and qualification steps.

Why are GlobalFoundries and RAAAM developing GCRAM memory on the FDX platform?

The companies state that on-chip memory has become a bottleneck as edge AI, automotive and IoT devices need more data capacity. They aim for a denser, lower-power alternative to SRAM, allowing increased on-chip memory or reduced silicon area while extending data retention and saving energy.

How is NXP Semiconductors involved in the GCRAM development with GlobalFoundries (GFS)?

NXP Semiconductors is not a development partner but is closely evaluating the benefits of the GCRAM solution. NXP views the projected area shrink and power reductions as offering a path to higher on-chip memory capacities and reduced off-chip data movement, which can improve system-level efficiency.

What applications are targeted by the GCRAM technology co-developed by GlobalFoundries (GFS) and RAAAM?

The GCRAM technology on GF’s FDX FD-SOI platform targets edge AI, automotive and IoT devices. The companies highlight these segments because they require large on-chip memory capacity, low power consumption and improved system-level efficiency.