Razer is making a clear move deeper into artificial intelligence, and it’s starting with new hardware built for developers who want serious AI performance without being tied to a full workstation. At CES 2026, the company revealed its first-generation compact AI accelerator device created in partnership with Tenstorrent, aiming to bring portable “edge AI” compute to everyday Thunderbolt and USB-equipped systems.
What makes this launch especially notable is the technology inside: the accelerator is powered by Tenstorrent’s Wormhole n150 and is based on the RISC-V ecosystem, pushing a more open and developer-friendly approach compared to many traditional consumer AI solutions. The goal is straightforward—give developers a modular, desktop-ready AI accelerator that can travel easily, connect quickly, and scale up when workloads demand more power.
The device is designed to plug into any system with Thunderbolt 5, Thunderbolt 4, or USB4, turning a compatible laptop or desktop into a more capable platform for running AI workloads locally. That includes popular generative AI tasks like large language models (LLMs), image generation models, and a broad mix of AI/ML experimentation and development. By focusing on local compute, this kind of accelerator can help developers prototype and test AI tools without relying entirely on cloud resources, while also supporting low-latency edge deployments.
A key selling point is scalability. Instead of locking users into one fixed performance tier, the compact accelerator can be expanded by daisy-chaining multiple units—up to four devices—creating a small desktop-sized cluster for larger models and heavier experimentation. For developers pushing bigger workloads, that modular approach can be an appealing middle ground between a single add-on device and a much more expensive dedicated AI server.
Razer and Tenstorrent are also emphasizing software flexibility. The accelerator works with Tenstorrent’s open-source software stack, which is intended to make it easier for developers to build, test, and iterate across a range of AI projects while tapping into Tenstorrent’s broader ecosystem.
According to Razer’s Travis Furst, Head of the Notebook and Accessories Division, the collaboration is about delivering the power, flexibility, and mobility edge AI developers want—combining Tenstorrent’s AI acceleration technology with Razer’s experience in high-performance engineering and external enclosure design. In other words, this isn’t just a concept device—it’s positioned as a practical tool for developers who want portable AI compute that can scale with their needs.
With this compact RISC-V-based AI accelerator, Razer is signaling that its AI ambitions go beyond software features and into purpose-built hardware—especially for creators and developers who want generative AI performance at the desktop, on the go, and closer to where data is actually used.
