MOUNTAIN VIEW, Calif., April 2, 2026 — Google has introduced Gemma 4, a new family of open models with open weights that is clearly aimed at bringing more capable AI onto local hardware. Released under the Apache 2.0 license, the Gemma 4 family includes four sizes: E2B, E4B, 26B A4B MoE and 31B Dense. Google […]
Google Pushes Multimodal AI Further Onto Edge Devices with Gemma 4 Read More +
We’re excited to announce Intel’s strategic partnership with Google to deliver optimized Gemma 4 models on Intel hardware from day one. This collaboration enables developers to leverage the power of Google’s latest AI models on Intel hardware: Intel® Core™ Ultra processors, Intel® Xeon® CPUs, and Intel® Arc™ GPUs. Developers can create AI applications that run
Gemma 4 Models Optimized for Intel Hardware: Enabling Instant Deployment from Day Zero Read More +
OSRAM utilizes the Quantus SNA workflow for high-precision silicon. In modern semiconductor design, Heterogeneous Integration is the new frontier. When light detectors for Medical Imaging or depth sensors for Autonomous Systems are packed onto a single die alongside radio systems, the risk of interference skyrockets. For ams OSRAM, “sensing the power of light” means more
See How ams OSRAM Revolutionize Optical Solutions with the Help of Cadence Tools Read More +
This blog post was originally published at Quadric’s website. It is reprinted here with the permission of Quadric. After years of cloud-centric inference, AI is moving to the edge. The “Goldilocks zone” of 3B to 30B parameter models is delivering GPT-4-class performance on smartphones, automotive systems, and industrial equipment — and creating an acute challenge for
The On-Device LLM Revolution: Why 3B-30B Models Are Moving to the Edge Read More +
The difference in optical efficiency at wider FOV for the same technology type for several commercial waveguides. Data was normalized to the narrow FOV product, with arrows representing the reduction in efficiency for wide FOV waveguides developed by the same company. A clear reduction in optical efficiency for waveguides of the same technology at wider
Evaluating Waveguide Technologies for AR Smart Glasses Read More +
This blog post was originally published at Samsung Semiconductor’s website. It is reprinted here with the permission of Samsung Semiconductor. Until recently, the evolution of mobile cameras has been centered on the image sensor and the image signal processor (ISP). In this conventional architecture, the image sensor converts light into electrical signals, while the ISP corrects
Inside the Intelligent Mobile Camera Powered by Exynos 2600 VPS Read More +
On April 15, 2026, at 9:00 pm PDT (12:00 pm EDT) NVIDIA will deliver a webinar “Unlock Real-Time Physical AI for the Industrial Edge” From the event page: Join us to learn how IGX Thor’s Blackwell-powered architecture is powering autonomous robots, surgical systems, and high-performance industrial automation at the edge. NVIDIA experts will walk through
Upcoming Webinar on NVIDIA IGX Thor Read More +
On April 16, 2026, at 1:00 pm EDT (10:00 am PDT) Boston.AI will deliver a webinar “Remembering to Forget: Agentic Memory Systems and Context Constraints” From the event page: As AI agents evolve from stateless responders into persistent, goal-directed systems, memory has become a central design challenge. The question is no longer just what agents
Upcoming Webinar on Agentic Memory Systems Read More +
At ISC West 2026 Qualcomm Technologies showcases its vision for the future of smart camera and security Key Takeaways: End-to-end AI development tools and the Qualcomm Insight Platform to enable customers to develop AI features once and scale across their entire product portfolio. Qualcomm Technologies’ camera solutions span video security, law enforcement and enterprise body
Microchip presents a customizable, target-agnostic solution to program wake words and voice commands. The ML model, generated and tested using a custom application, has low latency and a minimal memory footprint, making it ideal for resource-constrained embedded systems. The ML model can be integrated into voice-based applications running on any 32-bit microcontroller or microprocessor running
Lightweight Keyword Spotting Solution from Microchip Read More +
Here you’ll find a wealth of practical technical insights and expert advice to help you bring AI and visual intelligence into your products without flying blind.
Berkeley Design Technology, Inc.
PO Box #4446
Walnut Creek, CA 94596
