GSI Technology (NASDAQ:GSIT) is positioning its Gemini-II artificial intelligence processor as a production-ready hardware platform while it works to expand software tools and complete proof-of-concept programs that could lead to early production by the end of 2027, according to Vice President of Sales Didier Lasserre.
Speaking at a Canaccord session following the company’s June-quarter report, Lasserre said GSI has progressed from introducing Gemini-II about 18 months ago to placing the hardware into production and securing third-party validation and government-backed engagements.
Compute-in-Memory Architecture
Lasserre said Gemini-II’s compute-in-memory architecture is intended to reduce the power consumed by the repeated movement of data between processors and external memory in conventional CPU and GPU designs. The processor loads a model or database into on-chip memory and performs calculations within the memory bit line, he said.
That approach is particularly suited to search and high-performance computing applications, according to Lasserre. He said the company’s next-generation product, Plato, will target large language models, vision-language models and other edge applications by increasing the bandwidth for data entering the chip.
Plato is expected to offer what Lasserre described as near data-center performance within a power budget of 2 watts to 10 watts. He said the company expects to complete the chip’s design and tape-out in the spring.
While GSI develops its AI business, its long-standing SRAM business remains an important source of support for company operations. Lasserre said GSI has shipped more than 140 million SRAM devices over roughly 30 years. The products support AI-related infrastructure indirectly through customers including Cadence, whose emulation systems are used in chip design, and KYEC, which provides burn-in services for GPUs.
Proof-of-Concept Programs and Software Roadmap
GSI is currently working on two announced proof-of-concept programs for Gemini-II: a Department of Defense-funded drone surveillance project and a Phase I Smart City project with a municipality in Taiwan.
In the drone surveillance program, Lasserre said the customer required a time-to-first-token result below three seconds and power consumption below 50 watts. He said GSI and drone partner G2 Tech demonstrated a result of 2.5 seconds at 30 watts in a June laboratory demonstration for the Defense Department.
The Taiwan Smart City project is initially designed to analyze video from 20 cameras and identify events as they occur, rather than simply recording footage for later review. Lasserre said a potential Phase II would expand the system to 80 cameras and add audio capabilities. A potential Phase III production deployment could involve between 2,000 and 6,000 cameras by the end of 2027.
He said one Gemini-II chip would support every four cameras in such a deployment, alongside an annual recurring software license for the application.
GSI expects to release an alpha version of its AI-assisted software development kit this fall to select customers, followed by a broader release next year. Lasserre said the software is central to allowing customers and systems integrators to develop applications without needing to work directly with the chip’s machine code.
According to Lasserre, an application for the drone proof of concept took approximately one person-year to develop before the AI-assisted SDK. The internal team’s early use of the new tool indicates that development time could be reduced to weeks, he said.
Defense Interest and Funding Position
Lasserre said defense could be an early market for GSI’s AI products because government entities have shown early interest and made financial commitments through Small Business Innovation Research, or SBIR, awards.
- GSI has won two SBIR awards with the Air Force Research Laboratory.
- The company has won two awards with the Space Development Agency.
- It has also won an award with the U.S. Army for a ruggedized edge server concept.
One active Space Development Agency award is funding radiation testing of a commercial Gemini-II part for potential space use. Lasserre said testing completed in June showed zero single-event latch-ups, though the company was still awaiting the third-party report. Total ionizing dose testing was expected to begin at the end of August or early September.
The Army-funded effort involves a ruggedized edge server that could support object detection or synthetic aperture radar imagery processing in field environments. Lasserre said the concept could ultimately become a product and future revenue source.
On funding, Lasserre said GSI had $77 million in cash and no debt. The company is burning roughly $4 million per quarter, he said, with an expected additional expense of several million dollars during the spring related to Plato tape-out. He said the company believes its cash position is sufficient to meet its near-term milestones.
Looking ahead, Lasserre said investors should watch progress in completing the drone and Smart City proof-of-concept projects, the SDK rollout and Plato’s tape-out. “It’s really more of an execution” challenge than a technology-validation challenge, he said.
About GSI Technology (NASDAQ:GSIT)
GSI Technology, Inc is a fabless semiconductor company specializing in the design and development of high-performance memory products. Headquartered in Sunnyvale, California, the company was founded in 1995 and has focused its efforts on content addressable memory (CAM) and high-speed SRAM (static random-access memory) solutions. As a publicly traded company listed on NASDAQ under the ticker GSIT, GSI Technology leverages advanced architectures to meet demanding data-processing requirements.
The company’s core product portfolio includes ternary CAM (TCAM) devices, binary CAM (BCAM) devices and high-speed synchronous SRAM.
-logo-1200x675.png){kind=logo}
