The core of the new architecture rests on the Cedar multimodal model and AI Station, a hardware hub powered by the NVIDIA Jetson T5000. By utilizing Google’s Gemma open model, Cedar enables devices to handle complex reasoning at the edge, while the AI Station allows multiple applications to share computing resources. This configuration reduces the need for dedicated high-performance hardware in every endpoint, keeping latency-sensitive data within the home environment.
In section Releases
SDMC Unveils Physical AI Home Architecture at IBC 2026
At IBC 2026 in Amsterdam, SDMC debuted a deployable AI home architecture that integrates cloud-edge intelligence with emerging Physical AI capabilities. Following its selection for Google Cloud’s Proactive Physical AI Acceleration Program, the company showcased how shared edge computing can power responsive, privacy-focused services across entertainment and home security networks.
SDMC also expanded its hardware lineup to demonstrate these capabilities in practice. Collaborating with Amlogic, the company introduced an AI Home Display and an AI Streaming Box, alongside a Google TV 4K Projector. These devices integrate with the AI Fiber Gateway to provide autonomous network optimization and proactive maintenance. The focus remains on enabling operators to scale smart home services without the friction of complex, hardware-heavy deployments, bridging the gap between digital understanding and real-world physical actions.
Comments (0)
No comments yet. Be the first!