Sugon 64-Thread Mobile Workstation: 16GB VRAM, 16.9mm Thickness

Sugon has introduced a mobile workstation that challenges the traditional trade-off between processing power and portability. This device matters because it packs 64 threads into a chassis thin enough to slip into a standard laptop bag, offering serious compute power for field professionals. Buyers who need to run large language models on the go now have a hardware option that fits in a briefcase.

Sugon unveils a 16.9mm thin mobile workstation with 64 threads and 16GB VRAM for local AI inference.

The company unveiled this machine as the world's first mobile workstation with 64 threads. It targets users who require heavy local AI inference capabilities without being tethered to a desktop setup. The design prioritizes a slim profile, measuring just 16.9mm in thickness while housing significant internal components.

Specifications

• Thread Count: 64 threads
• Thickness: 16.9mm
• VRAM: 16GB
• AI Inference Speed: 50 Tokens/s (35B MoE)

Under the hood, the system relies on a processor delivering 64 threads and a graphics subsystem with 16GB of video RAM. These specifications enable the device to handle substantial computational loads in a compact form factor. The specific CPU and GPU model numbers remain undisclosed in the current reports.

Sugon claims the system can run a 35-billion parameter Mixture of Experts (MoE) model locally at a speed of 50 tokens per second. The vendor states this AI inference performance is up to three to five times faster than mainstream industry products currently available. No pricing or release date has been confirmed for this hardware.

The announcement establishes a new baseline for mobile AI workstations by combining high thread counts with dedicated VRAM in a thin chassis. Readers should note that specific component details and shipping timelines are still pending further disclosure from the manufacturer.