SST memBrain SAGE IP is already field-tested and deployed in robust 40 nm and 28 nm foundry processes using production-ready SuperFlash memory.
AnalogAI is bringing a new wave of intelligence to edge devices through a fresh partnership with Silicon Storage Technology (SST), a subsidiary of Microchip Technology.
AnalogAI has selected SST’s memBrain Synaptic Analog Generative Engine (SAGE) neuromorphic hardware IP to fuel its next-generation edge AI processors.
By pairing AnalogAI’s proprietary algorithms which can train and run inference simultaneously right in the field with SST’s proven compute-in-memory technology, the companies are unlocking high-performance AI at under one watt of power.
This breakthrough is tailor-made for resource-constrained, real-world applications where every milliwatt matters, such as autonomous drones, self-driving vehicles, and humanoid robots that need to adapt to dynamic environments on the fly.
Inside the memBrain SAGE IP Integration
At the heart of AnalogAI’s new processors is SST’s production-proven SuperFlash technology, designed to deliver high-performance, non-volatile storage directly on the chip without needing bulky external memory.
The robust SST memBrain SAGE IP package comes fully loaded with:
- memBrain Tensor In-Memory Logic Element (TILE): Built on an optimized ESF3 bitcell capable of storing up to 8 bits per cell (bpc) at nanoamp levels, paired with custom arrays, decoders, and driver circuitry.
- Precision Conversion & Logic: Features optimized DACs and ADCs, a summator, high-voltage bias circuitry, and nanoamp-level bitcell control logic.
- Support & Simulation: Backed by proprietary test circuitry, complete technical documentation, simulation models, and full IP integration support.
Scaled for the Future of Edge Computing
SST memBrain SAGE IP is already field-tested and deployed in robust 40 nm and 28 nm foundry processes using production-ready SuperFlash memory.
Looking ahead, the technology roadmap includes the active development of a 22 nm memBrain SAGE IP variant.
By combining AnalogAI’s real-world learning algorithms with SST’s ultra-low-power analog compute-in-memory (aCIM) architecture, the collaboration paves the way for smarter, more autonomous machines that can operate efficiently at the extreme edge.
Leadership Comments
“AnalogAI is pursuing innovative co-optimized solutions for edge AI inference that enable edge devices to adapt real-time to sudden changes in the environment,” said Mark Reiten, senior vice president of Microchip’s Intelligent Compute business unit. “As the core inference engine for AnalogAI’s first products, SST memBrain SAGE IP delivers the 
necessary compute performance coupled with the power efficiency AnalogAI requires to meet their application targets. We are pleased to welcome AnalogAI as the newest licensee of our SST memBrain IP, joining our rapidly expanding ecosystem.”
“AnalogAI selected SST’s memBrain SAGE IP after an industry-wide search of available offerings and determined the silicon-proven memBrain SAGE IP best enabled us to accelerate our development time while still achieving the ultra-low-power and high performance required in our market segment,” said Jaejun Lee, AnalogAI’s chief executive officer. “Analog Compute-in-Memory is a rapidly emerging field, and at AnalogAI we are pioneering new functionality for edge AI devices requiring real-world adaptation.”
Pricing and Availability
Customers interested in SST’s memBrain IP solutions and SuperFlash technology should access the SST website or contact a regional SST sales executive for details.
Those interested in AnalogAI’s products should visit the AnalogAI website or contact the AnalogAI team at contact@analog-ai.com.
