Video: Charting the Future of AI/ML with Open Standards and Global Collaboration
Philipp Tomsich, Technical Steering Committee Vice Chair, discusses RISC-V’s critical role in AI development, highlighting its flexibility for custom silicon design, strong ecosystem support, and emerging standards that enable efficient, domain-specific acceleration for AI applications, positioning RISC-V as the preferred architecture for building AI accelerators
Common Language for the Development of AI Systems
A New Standard For AI
RISC-V is quickly becoming the preferred standard for building performant and efficient AI accelerators, bringing together experts from different geographies and
industries to define the AI solutions of tomorrow.
A Common Language For AI
RISC-V provides a common language for AI development, as an industry standard ISA, backed by a cohesive ecosystem for AI/ML development addressing all market segments.
Ecosystem Support
A unified ecosystem unites global experts across industries to define tomorrow’s AI solutions, delivering innovative technologies for future generations.
Software-Based Approach to Hardware
Software Focused
The extensible industry standard RISC-V ISA enables a software-focused approach to AI hardware, and a unified programming model across AI workloads running on CPU, GPU & NPU
Workload-Based Customization
RISC-V’s modular architecture enables industry leading differentiation, through the development of custom instructions and accelerators targeted at your software work–load
Future Proofed
The latest advancements in AI/ML algorithms can be quickly integrated into hardware designs, keeping pace with fast-evolving demands.
Control your Compute Roadmap & Supply Chain
Tailored Customization
RISC-V allows for domain-specific customization tailored to particular applications and workloads by enabling the selection of appropriate ratified extensions from the standard.
Extensibility
RISC-V enables the addition of ratified extensions from the standard or vendor-
developed extensions to differentiate products with application-specific functionality.
Freedom to Innovate
A standardized yet flexible platform allows designers to rapidly integrate cutting-edge research into hardware without being hampered by proprietary constraints.
Building a Better Business Case for AI
Cost Efficient
RISC-V enables rapid innovation and cost-effective development.
No Lock-In
Free from restrictions of vendor lock-in RISC-V enables open collaboration and pooled resources to advance one interoperable global standard.
Flexibility
By reducing barriers to entry and democratizing AI, RISC-V is a standardized yet
flexible foundation that upends the economics of custom silicon x enabling industry-leading differentiation.
Hear from our Members About RISC-V and AI
RACE: Powering Next-Gen RISC-V AI Solutions
Shan Lui, Beijing Institute of Open-Source Chip (BOSC)
LLM Inference on RISC-V Embedded CPUs
Yueh-Feng Lee, Andes Technology
The Future of AI and Security
Andrew Dellow, Qualcomm; Kris Murphy, NVIDIA; Pete Bernard, tinyML Foundation; Pete Warden, Useful Sensors Inc; Andrea Gallo, RISC-V International
Lessons Learned in Using RISC-V for Generative AI and Where We Can Go from Here
Jayesh Iyer & Josep M Perez, Esperanto Technologies
Building Tool Chains for RISC-V AI Accelerators
Jeremy Bennett, Embecosm
The Benefits of Building New AI Accelerators with RISC-V
Cliff Young & Martin Maas, Google DeepMind
RISC-V at NVIDIA: One Architecture, Dozens of Applications, Billions of Processors
Frans Sijstermans, Vice President Multimedia Arch/ASIC, NVIDIA
All-in-One RISC-V AI Compute Engine
Roger Espasa, Semidynamics
Panel Discussion: Accelerating AI Innovation with RISC-V
Recent AI Stories on the RISC-V Blog
Channel Life: The IoT semiconductor market is heading into a major shift by 2026, driven by the mainstream adoption of edge AI, the rise of open and modular architectures like RISC-V and chiplets, and increasing…
Semiconductor Engineering: AI continues to migrate towards the edge and is no longer confined to the data center. Edge AI brings several key advantages, delivering intelligence closer to where data is generated, improving latency for…
Embecosm used the oneAPI Construction Kit to explore accelerating PyTorch using RISC-V cores, trying over a thousand in emulation and some on an FPGA. This case study shows how the oneAPI Construction Kit lets you…
Andes: d-Matrix and Andes have partnered to integrate Andes’ high-performance AX46MPV RISC-V CPU IP into d-Matrix’s next-generation Raptor accelerator, the first to feature 3D In-Memory Compute (3DIMC) technology for faster, more efficient AI inference at…
Developed with SiFive, the dual-core SoC uses patented error correction to achieve a record 16.8 µW/MHz/DMIPS, targeting next-gen wearables, drones, and IoT sensors. In a move aimed at the growing market for intelligent edge devices,…
Google Research has open-sourced its Coral NPU IP (previously codenamed Kelvin), which it is giving to the industry in a bid to accelerate edge AI implementations by reducing fragmentation and improving security. Synaptics is the…