NVIDIA CUDA Expands Support to RISC-V Processors in AI Applications, Challenging x86 and ARM Dominance

NVIDIA has achieved a significant breakthrough by extending its CUDA support to RISC-V processors, a development that heralds new opportunities in the artificial intelligence (AI) sector.

NVIDIA CUDA Now Compatible with RISC-V Processors

For several years, the AI landscape has been heavily dependent on data center CPUs built on either the x86 or ARM architectures. Intel and AMD continue to dominate the x86 market, while NVIDIA and several tech giants have carved out niche solutions within the ARM framework. However, this scenario is set for a transformation. In a recent update, RISC-V announced that its architecture will now support NVIDIA’s CUDA software stack via a porting mechanism, thus igniting potential market expansion.

Exciting news from #RISCVSummitChina, as Frans Sijstermans from NVIDIA announces CUDA is coming to RISC-V! This port will enable a RISC-V CPU to be the main application processor in a CUDA-based AI system.#RISCV #RISCVEverywhere pic.twitter.com/08C2ghPHq9

— RISC-V International (@risc_v) July 18, 2025

NVIDIA’s CUDA is widely regarded as a leader in AI computation, and with RISC-V poised to support this influential platform, we can expect a surge in its adoption across diverse industries. One of the primary advantages of the RISC-V architecture lies in its open-source nature, enabling developers and businesses to utilize, modify, implement, and distribute the Instruction Set Architecture (ISA) without incurring licensing fees. This open access is likely to spur greater adoption among smaller businesses and startups.

Blue circuit board featuring chips labeled WORMHOLE by Tenstorrent on a gray background. — Tenstorrent RISC-V AI chip

Moreover, RISC-V is recognized for its scalability advantages. Its minimalist instruction set simplifies chip design and verification, thereby accelerating the development and testing phases. The architecture is particularly promising for edge AI applications, especially given the current dominance of ARM and x86 in large-scale clusters. Although RISC-V has yet to make substantial inroads into the AI sector, one notable company leading the charge is Jim Keller’s Tenstorrent.

Tenstorrent is focused on delivering powerful and cost-efficient AI processors, with its Wormhole AI series comprising models such as the Wormhole n150 and n300. Additionally, RISC-V’s open-source appeal significantly resonates with Chinese developers. With the integration of NVIDIA’s CUDA support, we are likely to see a marked increase in interest within the AI domain.

Source & Images