Razer is stepping deeper into the world of professional AI hardware with the announcement of its Forge AI Dev Workstation, a custom-built desktop aimed squarely at developers, researchers, and teams building and testing large AI models. Revealed at CES 2026, the system is designed to make high-end AI compute easier to buy and deploy—without the headache of planning a complex multi-GPU workstation build from scratch.
The core appeal is simple: run large-language models locally instead of relying on rented cloud servers. For many AI developers, doing more work on-premises can mean better control over sensitive data, fewer ongoing usage bills, and the freedom to iterate without worrying about hourly compute costs.
To support demanding AI workloads, the Forge AI Dev Workstation can be configured with Intel Xeon W or AMD Threadripper Pro processors, scaling up to 96 CPU cores. On the graphics side, Razer offers options that reach up to four professional-grade GPUs, including Nvidia RTX Pro or AMD Radeon Pro models. This type of multi-GPU setup is often used for training, fine-tuning, and running large AI models, as well as for other compute-heavy work like simulation and advanced rendering.
Memory and storage expansion are also a major part of the design. The workstation supports up to eight DDR RDIMM slots, along with storage configurations that include up to four PCIe Gen5 M.2 NVMe SSDs and as many as eight SATA drives. Power delivery is handled by a hefty 2,000W power supply—important for stability when running multiple GPUs at sustained loads. Cooling comes via a high-airflow layout with four 120 mm fans positioned at the front and back to help move air efficiently through the system.
For teams that need more compute than a single tower can provide, Razer also highlights scalability. Each workstation includes dual 10-Gigabit Ethernet ports, making it possible to network multiple units together into a local compute cluster. There’s also an option to rack multiple systems for better space efficiency in lab or office environments.
Because Forge AI Dev Workstations are customized, pricing and exact configurations depend on the build. Razer is asking interested buyers to request a quote and detailed specifications based on their needs.
It’s also worth noting that not everyone needs (or can justify) a top-tier AI workstation to get started with AI development. Many smaller or “distilled” AI models can run on more modest hardware, which can be a practical approach for learners, students, and developers prototyping on mainstream laptops or midrange GPUs before scaling up to a dedicated multi-GPU machine.
