-- At this year’s ATxSG technology festival, one piece of hardware quietly stole the spotlight from the usual cloud giants. Nanyang Singtech, a deep-tech firm incubated at Nanyang Technological University (NTU), showcased the SingNova-H Studio—a compact, next-generation AI PC. For global businesses and developers eager to escape cloud dependency, 128 GB of unified memory and 200 TOPS of computing power make this configuration nearly perfect for running large models locally.
The global AI economy is booming, yet most of its computing power still lives in distant data centers. That model works fine—until it does not. Until latency spikes, subscription fees climb, or regulators start asking where sensitive data actually sits. Nanyang Singtech’s argument is simple: what if powerful AI computing sat right on your desk, kept your data in the room, and spoke the language of open hardware?
What This Machine Can Actually Do
The SingNova-H Studio is built around a RISC-V dataflow processor, different from the familiar chips inside most laptops and servers today. No need to dive deep into silicon design—the core idea is handling information more efficiently by cutting down the endless back-and-forth between memory and computing cores. The result, says the company, is a small box that punches far above its weight.
Inside sits a 12-core RISC-V CPU and an AI engine rated at 200 TOPS for integer work and 32 TFLOPS at half-precision. In plain language, that means the machine can run large AI models with up to 70 billion parameters entirely on the device. To put that in context, many of today’s writing, coding, and analysis models operate at roughly that scale. Everything happens locally, so documents, voice recordings, or camera feeds never need to leave the building to be processed.
The system also carries up to 128 GB of error-correcting memory, fast NVMe storage, and dual 10-gigabit Ethernet ports. It can drive multiple monitors via HDMI, DisplayPort, and USB-C, and under normal use consumes about as much power as a bright desk lamp. Despite the muscle, typical draw stays between 10 and 75 watts.
Why Local Processing Matters
For financial institutions, healthcare providers, and government agencies across Asia and around the world, local processing is no longer just a nice-to-have—it is becoming a compliance necessity. Data-residency regulations are tightening globally. Shipping customer records overseas for AI analysis adds cost and legal headaches. A workstation that can handle advanced inference in-house bridges the gap between underpowered consumer laptops and pricey cloud contracts.
Beyond big business, the impact reaches across industries. Manufacturers testing visual quality control, studios rendering AI-generated images, and universities teaching open-source architecture all share one problem: getting capable hardware at a reasonable price. The SingNova-H Studio is positioning itself as an option that needs neither a server room nor a foreign cloud account.
The machine runs Linux and other open-source operating systems, and supports the mainstream AI frameworks developers already use. That makes it easier for engineering teams to migrate from other platforms.
From Campus Lab to Show Floor
Nanyang Singtech’s NTU roots matter. Singapore has long cultivated a pipeline from university research to commercial product, especially in semiconductors and AI. By building its flagship AI PC on the open RISC-V standard instead of proprietary architectures, the company is also betting on an open, royalty-free ecosystem that gives developers and hardware makers more freedom to innovate without licensing restrictions.
The company’s wider portfolio covers intelligent computing, smart finance, energy systems, and autonomous driving. This AI PC is essentially the desktop version of that same philosophy: compute, storage, and intelligence packaged for direct use, rather than rented by the hour from a distant data center.
The Bigger Picture
If the SingNova-H Studio lives up to its promise, it could help nudge a subtle but important shift in how organizations worldwide use AI. Instead of automatically sending every heavy workload to an overseas cloud provider, companies may start keeping more intelligence at the edge—on factory floors, in editing suites, or inside research labs.
For a young company grown from local soil, that is an ambitious vision. But in an industry determined to own more of its technology stack, ambition may be exactly what the world is waiting for.
About Nanyang Singtech
Nanyang Singtech is a Singapore-based deep technology company incubated by Nanyang Technological University. It focuses on intelligent computing, smart finance, intelligent energy, and embodied intelligence, delivering full-scenario intelligent infrastructure covering intelligent computing, storage, energy, and autonomous driving systems.
Contact Info:
Name: LONG HAIYAN
Email: Send Email
Organization: NANYANG SINGTECH PTE. LTD.
Website: https://www.nanyang-singtech.com/
Release ID: 89192440
Google
RSS