From All-in-One IP to Cervell™: How Semidynamics Reimagined AI Compute with RISC-V
In an era where artificial intelligence workloads are growing in scale, complexity, and diversity, chipmakers are facing increasing pressure to deliver solutions that are not only fast, but also flexible and programmable. Semidynamics recently announced Cervell™, a fully programmable Neural Processing Unit (NPU) designed to handle scalable AI compute from the edge to the datacenter. Cervell represents a fundamental shift in how AI processors are conceived and deployed. It is the culmination of Semidynamics’ IP offerings evolution from modular IP components to a tightly integrated, unified architecture, rooted in the open RISC-V ecosystem.
Key Takeaways
- Cervell™ is a fully programmable Neural Processing Unit (NPU) designed for scalable AI compute from the edge to the datacenter.
- Cervell integrates CPU, vector units, and tensor engines into a single processing entity, eliminating the need for external CPUs and reducing performance bottlenecks.
- The architecture is based on the open RISC-V ecosystem, allowing for deep customization and flexibility in processor design.
To read the full article, click here
Related Semiconductor IP
- All-In-One RISC-V NPU
- NPU
- NPU IP Core for Mobile
- NPU IP Core for Edge
- Specialized Video Processing NPU IP
Related Blogs
- Custom Compute for Edge AI: Accelerating innovation with Lund University and Codasip University Program
- Semidynamics – Customizable RISC-V Technology for the Next Five AI Revolutions
- RISC-V customization, HW/SW co-optimization, and custom compute
- Empowering AI-Enabled Systems with MIPI C/D-PHY Combo IP: The Complete Audio-Visual Subsystem and AI
Latest Blogs
- Breaking the Silence: What Is SoundWire‑I3S and Why It Matters
- What It Will Take to Build a Resilient Automotive Compute Ecosystem
- The Blind Spot of Semiconductor IP Sales
- Scalable I/O Virtualization: A Deep Dive into PCIe’s Next Gen Virtualization
- UEC-LLR: The Future of Loss Recovery in Ethernet for AI and HPC