AMD’s Ryzen AI Embedded P100 Sequence processor. | Supply: AMD
AMD in the present day expanded its Ryzen AI Embedded P100 Sequence processor portfolio. The corporate stated the expanded line will higher swimsuit the quickly evolving computing wants of manufacturing unit automation, cell robots, and different AI-driven edge purposes.
The Santa Clara, Calif.-based firm stated its new processors function as much as two occasions greater CPU core models, eight occasions greater graphics processing unit (GPU) compute, and an estimated 36% greater system tera operations.
On a single chip, the processors function:
- Eight to 12 “Zen 5” cores
- As much as 80 system TOPS for bodily AI acceleration
- AMD RDNA 3.5 graphics for real-time visualization
- A neural processing unit (NPU) based mostly on the AMD XDNA 2 structure for low-latency, power-efficient AI inference
AMD Ryzen AI Embedded P100 Sequence processors that includes eight to 12 cores are at the moment sampling, with manufacturing shipments anticipated to start in July 2026. P100 Sequence four- to six-core processors are sampling now, with manufacturing anticipated within the second quarter of 2026.
AMD affords scalable AI compute
The processors allow consolidation of programmable logic controllers (PLCs), machine imaginative and prescient, and human-machine interface (HMI) right into a single industrial PC, whereas delivering the CPU efficiency required for real-time inspection and course of optimization. The built-in GPU and NPU speed up multicamera imaginative and prescient and wealthy HMI dashboards whereas enabling low-latency anomaly detection utilizing fashions like DeepSORT, RAFT-Stereo, CenterPoint, GDR-Web, PaDiM, and Llama 3.2-Imaginative and prescient.
For cell robots, the processors handle navigation, movement management, and route planning on the CPU. In the meantime, the GPU processes multicamera feeds for spatial consciousness, Visible SLAM, and superior AI workloads like vision-language-action (VLA) fashions. Unified reminiscence between the CPU and GPU unlocks low latency for higher responsiveness.
The NPU delivers always-on low-power inference for object detection and scene understanding utilizing fashions equivalent to YOLOv12 and MobileSAM.
The processors allow the powering of 3D imaging for ultrasounds, endoscopes, tissue classification, and tumor detection on the edge utilizing fashions like U-Web, nnU-Web, and MONAI. The processors speed up image-to-report workflows with MedSigLIP and help scientific reasoning and Q&A with Med-PaLM 2. Healthcare authentic tools producers (OEMs) can consolidate imaging, AI evaluation, and reporting on a scalable, long-life-cycle x86 embedded platform.
In contrast with the prior technology AMD Ryzen Embedded 8000 Sequence, the P100 Sequence is predicted to offer as much as 39% greater multithreaded efficiency and as much as 2.1 occasions greater whole system TOPS2. The brand new processors ship distinctive AI performance-per-watt and help virtually twice the variety of digital machines and bigger massive language fashions (LLMs), like Llama3.2-Imaginative and prescient 11B, than the present P100 Sequence to allow extra superior AI and combined workloads.
ROCm software program help and virtualized reference stack out there
Assist for the AMD ROCm open software program ecosystem brings a confirmed, open-source AI software program stack to embedded purposes. Builders can run customary AI frameworks whereas counting on open-source compilers, runtimes, and libraries – all whereas having quick entry to embedded-ready fashions with out rewriting code.
On the programming stage, ROCm software program makes use of the open-source Heterogeneous-computing Interface for Portability (HIP), decoupling GPU programming from the {hardware} and eliminating vendor lock-in between the software program stack and the {hardware}.
The tightly built-in CPU, GPU, and NPU structure permits environment friendly workload partitioning and predictable latency beneath combined workloads, whereas the usage of acquainted frameworks and software program stacks helps simplify and streamline improvement and deployment throughout broad use circumstances. This stage of integration permits superior compute and graphics capabilities with out the necessity for extra exterior elements, making it simpler for OEMs and system integrators to design scalable platforms.
AMD “Zen 5” CPU cores present the isolation and efficiency headroom to consolidate a number of crucial workloads on a single platform with deterministic, multitasking habits. Moreover, AMD delivers a packaged and vertically built-in virtualized reference stack for industrial mixed-criticality purposes.
Constructed on the Xen hypervisor, it runs Linux, Home windows, Ubuntu, and RTOS environments in remoted domains to ship security, real-time efficiency, and adaptability.
The put up AMD expands Ryzen AI processor product line appeared first on The Robotic Report.
