Sugon, a Chinese supercomputer manufacturer, has teased a new mobile workstation (MWS) designed for AI performance. The device integrates a domestic 64-thread processor and a GPU with 16GB of VRAM. This configuration allows it to achieve a performance of 50 tokens per second when running local 35B MoE AI models.
The mobile workstation features a thin design, measuring 0.665 inches (16.9 mm) thick. This places it within the category of thin and lightweight premium laptops, distinguishing it from bulkier portable workstations previously seen.
While Sugon did not officially disclose the specific processor model, it is expected to be sourced from Hygon, a long-time partner and company in which Sugon is the largest shareholder. Hygon has previously offered processors with up to 32 cores and 64 threads, including four generations (C86-1G to C86-4G series) that utilized licensed Zen IP from AMD. The workstation might also use Hygon's newer C86-5G chips, which reportedly feature a new self-developed microarchitecture.
This announcement highlights ongoing efforts within China to develop domestic hardware solutions for advanced computing, particularly in the AI sector. The integration of a powerful domestic CPU and significant VRAM in a portable form factor addresses the growing demand for on-device AI processing capabilities.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's briefNew every morning, and the back catalogue is archived by date.
Chinese supercomputer manufacturer Sugon has revealed an upcoming mobile workstation featuring a domestic 64-thread processor and 16GB of VRAM, capable of processing 50 tokens per second on local 35B MoE AI models. This development signifies advancements in China's domestic hardware capabilities for AI applications, offering a portable solution for demanding computational tasks.