Beyond PCIe Compliance: Why Stress Testing Is Crucial For Edge AI Deployments


Passing PCI Express (PCIe) compliance is different from being ready for the field. A PCIe link can clear every test in a controlled lab environment and still develop margin problems six months into deployment. That’s because a compliance traffic generator isn’t designed to replicate real-world operating conditions, such as thermal stress, electrical noise, and the kind of bursty inference t... » read more

The Edge LLM Offload Story


By Karthikeyan Shanmuga Vadivel and Sauryadeep Pal Developers and system architects today face a growing demand to enable large language model variants on device. They are facing pressure to support transformer-capable models on constrained devices to ensure data privacy, eliminate cloud API charges, and provide offline reliability. On-device execution is also becoming a necessity to meet st... » read more

Connectivity and Compute in Next-Gen Edge Devices


AI-native Edge devices are reshaping IoT by converging AI, connectivity and compute into a single platform. This paper highlights how Synaptics SYN765x brings Wi-Fi 7, local AI processing and intelligent sensing together to reduce latency, lower costs, enhance privacy and accelerate next-generation connected device design. Read more here. » read more

Beating the Edge AI Power Wall with Low Voltage Foundation IP


Edge AI is pushing the limits of power efficiency as intelligence moves closer to the data source. Designing for ultra-low voltage operation is now essential to achieve optimal performance-per-watt—but it introduces significant complexity in modeling, variation, and design predictability. In this white paper, discover how a unified, silicon-proven Foundation IP platform approach enables relia... » read more

Flexible AI-MCU For Fast Inference of Transformer Models At The Ultra-Low-Power Edge (ETH Zurich, U. Bologna)


Researchers from ETH Zurich and University of Bologna have released “CHIMERA: A Flexible and Scalable 3.1 TOPS/W AI-MCU with Transformer Accelerator and 563 Gb/s Shared-L2 Memory Subsystem with QoS Guarantees”. Abstract “We present Chimera, a flexible and scalable Microcontroller Unit (MCU) designed to accelerate real-time inference of rapidly evolving transformer-based models a... » read more

Building Fixed HW Implementations of Neural Networks (Yale, Cornell et al.)


Researchers from Yale University, Cornell University, Boston University, and NTT Research have published “Physical Foundation Models: Fixed hardware implementations of large-scale neural networks”. Abstract "Foundation models are deep neural networks (such as GPT-5, Gemini~3, and Opus~4) trained on large datasets that can perform diverse downstream tasks -- text and code generation, q... » read more

Why Vision LLMs Force A Rethink Of Edge AI Hardware


As vision-centric large language models move on-device, performance measured in raw TOPS is no longer enough. Architectures need to be built around real workloads, memory behavior, and sustained utilization, especially at the edge. Vision LLMs are changing the edge AI equation For the last decade, most edge AI silicon has been built to do one job extremely well: run convolutional networks for... » read more

Vision-Language-Action Models Arrive


The AI model type capturing the most attention across robotics and autonomous vehicles right now is the vision-language-action model, or VLA. At embedded AI conferences this year, particularly the recently held Embedded Vision Summit, VLAs were a main topic of discussion – not as a research curiosity, but as the architecture that teams building autonomous systems are actively targeting. If yo... » read more

Beyond the Clinic: A Blueprint For Developing Reliable, Edge AI-Enabled Medical Devices


In a quiet farmhouse in rural Utah, hundreds of miles from the nearest city, a pregnant mother wakes up and waits for a kick that doesn’t come. In this part of the country—one of the many "medical deserts" where 30% of counties lack a single gynecologist—the nearest hospital is a 500-mile journey. Usually, this moment of silence leads to a desperate phone call to an HMO where a nurse asks... » read more

Designing Chips In The Context Of Rapidly Evolving AI


Key Takeaways: Agentic edge AI drives long-lived, tool-mediated loops with variable demands for compute, tokens, and memory. Edge PPA is dominated by memory hierarchy and data movement, forcing tight feature triage and robust RAS. Rapid model churn (multimodal, MoE, new formats) requires programmable, headroom-rich compute, interconnect, and runtime. Experts At The Table: Ch... » read more

← Older posts Newer posts →