AMD unveils Kria module for real-time control, unified memory for robots
AMD has launched the Ryzen AI Embedded X100 and Kria AI SoM, bringing unified memory and real-time control to challenge NVIDIA in robotics. The post AMD unveils Kria module for real-time control, unified memory for robots appeared first on The Robot Report.
Overview
The Kria module is designed to deliver compute and AI workload processing for physical AI. | Credit: AMD
Advanced Micro Devices Inc. is making a full-stack play for robotics with its new Ryzen AI Embedded X100 series and Kria AI Robotics platform. AMD today promised deterministic real-time control, unified CPU–GPU–NPU memory, and an open, non–vendor-locked software stack that it claimed can outperform NVIDIA Orin and Thor on system-level robotics workloads.
Built on a unified memory architecture that reduces latency across CPU, GPU, and NPU, the Ryzen AI Embedded X100 line targets “firm” and “hard” real-time control via BIOS and Linux optimizations, AMD QoS (quality of service) features, and Zen-based virtualization.
On top of the silicon, AMD is rolling out the Kria AI system-on-module and Robotics Development Kit, the open-source AMD Robotics Sophie Suite software stack, and a curated Robotics Partner Network. The Santa Clara, Calif.-based company said they position it to support everything from industrial arms and autonomous mobile robots (AMRs) to humanoids with a single, scalable platform.
AMD launches Ryzen AI embedded X100 processor
AMD earlier this month announced the new Ryzen AI Embedded X100 family of system-on-a-chip (SoC) components.
The new chips deliver a unified CPU–GPU–NPU architecture with a focus on embedded AI use cases, said the company. It designed the X100 for high-end robotic and edge AI applications based on its earlier sibling, the P100.
AMD delivered a “shot across the bow” of NVIDIA‘s chipset designed for physical AI applications. The X100 is now AMD’s flagship SoC for robotics applications.
The new Ryzen X100 is the flagship SoC for the AMD robotics lineup. | Credit: AMD
What X100 adds for robot builders
The X100 adds unified memory and heterogeneous compute for robotics workloads, and it explicitly reduces data copies. This is key for tasks such as perception, sensor fusion, and planning tasks, and unifies it all on a single platform.
Developers can deploy the X100 into a variety of firm or hard real-time applications. “Firm real time” is achieved via Linux + BIOS optimizations (interrupt latency target < 7 μs at six-nines). In addition, by using AMD QoS, the system can deliver L3 cache reservation, memory bandwidth allocation, and thread isolation.
For hard real-time use cases via virtualization, AMD recommended the Zen hypervisor, FreeRTOS virtual machine, cache coloring, and VM isolation. This is in contrast to traditional dual-device CPU with a discrete GPU architecture in a robot.
Save the date for RoboBusiness 2026Comparing X100 system-level performance with NVIDIA
NVIDIA is currently the dominant platform for physical AI. The release of the AMD X100 gives developers an alternative to NVIDIA architecture, while also putting price pressure on NVIDIA.
“In that market [aerospace and defense market, with a focus on signal-processing workloads], we’re actually delivering three times the FP32 compute performance relative to NVIDIA Thor,” said Rob Bauer, senior manager of product management and marketing for the x86 Embedded APU portfolio at AMD.
“They’re pretty highly optimized around inferencing in that architecture, right?” he said. “We’re still delivering floating-point signal processing performance here with the X100 series, so that allows us to do some pretty amazing things.”
AMD claimed that its Kria AI SoM delivers 3.4x better real-time performance over NVIDIA Thor T5000, has 1.6x spare compute capacity, and offers 2.3x more agentic AI capacity.
AMD X100 performance versus NVIDIA for robotics workloads. | Credit: AMD
AMD introduces the Kria AI SoM
Today, AMD announced the Kria AI SoM built on the X100 SoC. The architecture features a 1.5 μs control-loop closure and adoption of COM-HPC instead of a proprietary module.
Details
Compared with its consumer Ryzen products, AMD said the Kria/X100 line offers extended reliability, burn-in, test coverage, and fit rates necessary for an industrial application.
A new Kria AI robotics development kit is also being released to jump-start developers who want to test the new compute architecture.
The dev kit includes a robotics carrier card: GMSL, high-speed Ethernet, IMU, Wi-Fi/Bluetooth, etc. There is an integrated FPGA on the baseboard for real-time I/O, sensor fusion, and safety features.
There are also open-source baseboard schematics available to speed the development of production designs.
“What makes this [Kria] SoM differentiated is tying it back to the feedback [from customers],” said KV Thanjuvar Bhaaskar, robotics lead and senior manager at AMD. “They want open, non-vendor-lock-in. So what we did is adopt an open standard. Here, the standard we adopted is COM HPC. AMD can build it, and any other company on the planet can adopt it and build it. So we don’t want to have a proprietary form factor.”
The Kria system supports ROS 2 / Nav2 acceleration on X100, with a focus on deterministic control and compatibility with current robotics workflows.
AMD ROCm is supported as a common AI stack from cloud to robot and delivers robot-optimized AI pipelines.
AMD is positioning Kria as the foundation for robotics development. | Credit: AMD
AMD builds robotics partner network for Kria
AMD said it is committed to building a network of robotics partners, primarily sensor and software companies, to support the X100 and Kria Robotics SOM in their respective products. Some of these partners include Open Robotics, Open Navigation, and OpenCV, as well as sensor and original design manufacturer (ODM) partners such as analog/IMU vendors and system integrators.
The goal of the program is to focus on curated collaborations while supporting full robotics systems with the necessary sensing, compute, software, and integrations.
Early customers include Castec International, with its semiconductor fab AMR, and humanoid robot developer Foundation Robotics, which is migrating from Intel and NVIDIA to AMD X100 and future FPGA-based hand control.
The post AMD unveils Kria module for real-time control, unified memory for robots appeared first on The Robot Report.
Source
Originally published at www.therobotreport.com.
Related Articles
Source: https://www.therobotreport.com/amd-unveils-kria-module-real-time-control-unified-memory-robots/
