RISC-V Summit North America 2025 marked another exciting milestone for the global open hardware community—and for DeepComputing, it was a moment to showcase innovation, collaboration, and the growing power of RISC-V in the AI era.
Held in Santa Clara, the event brought together top silicon vendors, OS communities, and developers advancing the RISC-V ecosystem. DeepComputing’s booth and sessions became a focal point for discussions around RISC-V AI PCs, AI enablement, and open software stacks.
At the DeepComputing booth, visitors experienced firsthand the world’s first RISC-V AI PC, powered by ESWIN 7702RISC-V SoC, featuring 8 SiFive’s high-performance P550 CPU cores, and running Ubuntu Desktop 24.04 LTS.
The system delivers up to 50 TOPS of AI performance, enabling developers to run large language models (LLMs) locally. Demonstrations included VLC media player with local AI features, and browser-based AI applications, sparking deep discussions on open and sustainable AI computing.
The booth drew strong interest from developers, partners, and industry leaders exploring how RISC-V can redefine the future of personal and edge AI computing.
Matt St Onge, Associate Principal SolutionsArchitect from RedHatand Yuning Liang, Founder and CEO of DeepComputing, hosted a hands-on workshop titled “Getting Started with RISC-V, Linux and AI”, where participants built and deployed AI workloads directly on RISC-V platforms.
Using the open-source guide available on GitHub, developers learned how to:
The session was fully packed and received positive feedback for its practical, developer-focused approach—empowering participants to go beyond theory and get their hands on real RISC-V hardware.
Yuning delivered an inspiring talk, “Enabling Intelligent Media Playback on RISC-V – Running VLC”, showcasing how the VLC media player now runs smoothly on RISC-V architecture with AI extensions.
Using the DC-ROMA RISC-V AI PC, the demo illustrated real-time local AI enhancements such as speech recognition, translation, and subtitle generation—all powered natively on RISC-V without cloud dependency.
This presentation highlighted DeepComputing’s ongoing efforts to bring open media and AI experiences to users through sustainable, open hardware.
Kathy Grori, Advisor of DeepComputing, delivered a session titled “Running WebLLM in the Browser on RISC-V Toward Lightweight, Local AI Experiences”.
Her talk demonstrated how WebLLM enables developers to run language models directly in a web browser on RISC-V hardware—without the need for GPUs or external servers.
The audience was fascinated by how browser-based AI can deliver privacy-preserving, low-latency inference across RISC-V devices, further unlocking new potential for lightweight, local AI computing.
At the Summit, we officially announced our RISC-V Support Program, a new initiative designed to help developers, businesses, and ecosystem partners adopt and optimize RISC-V platforms.
Through this program, DeepComputing will provide:
This marks a major step forward in DeepComputing’s mission to empower global developers and accelerate the adoption of open computing.
From pioneering the RISC-V AI PC to driving software enablement and developer education, DeepComputing continues to push the boundaries of what’s possible with open hardware.
We thank everyone who visited our booth, joined our sessions, and shared in our vision. Together, we’re building a more open, intelligent, and sustainable computing future.
Follow us on LinkedIn, X, and Facebook for more updates, and visit our website to learn more about the RISC-V innovations.
