-- At this year’s ATxSG technology exhibition, Nanyang Singtech, a deep-tech company incubated by Nanyang Technological University (NTU), unveiled the SingNova-H Studio for the first time. This next-generation AI PC is powered by a RISC-V dataflow architecture system-on-chip (SoC). The debut marks a critical step for the company into full-scenario intelligent infrastructure, spanning intelligent computing, smart finance, intelligent energy, and embodied intelligence.
The SingNova-H Studio arrives at a pivotal moment for the AI industry. As enterprises and developers increasingly face cloud dependency, data privacy risks, and soaring inference costs, the market demand for powerful local AI computing has never been more urgent. Nanyang Singtech’s solution is a compact desktop workstation that delivers 128 GB of unified high-capacity memory and 200 TOPS of local large-model computing power—without ever sending data off-premises.
Breaking the Control-Flow Bottleneck
Traditional control-flow architectures have long suffered from the "memory wall"—the efficiency drain caused by repeatedly shuttling data between memory, cache, and processing units. The SingNova-H Studio fundamentally overcomes this limitation through a RISC-V dataflow architecture with control-data collaborative optimization. This design ensures high-concurrency, low-latency execution of video processing, AI inference, and scientific computing workloads on a unified platform.
At its core is a 12-core RISC-V CPU running at 2.0 GHz, integrated with an AI accelerator delivering 200 TOPS (INT8) and 32 TFLOPS (FP16). This throughput enables the workstation to run large language models with up to 70 billion parameters entirely on the device. All data remains local, delivering ultimate privacy and security while eliminating the network latency common in cloud-based inference.
Memory and Bandwidth for Real-World Workloads
Edge AI workloads are notoriously memory-hungry. The SingNova-H Studio features up to 128 GB of LPDDR5 ECC unified memory, providing ample room for massive model weights and high-resolution datasets. Storage comes via flexible NVMe PCIe 4.0 SSD configurations, while dual 10 GbE Ethernet ports deliver high-throughput connectivity for clustered deployments or data-intensive pipelines. The system also includes a PCIe 4.0 x16 expansion slot for discrete GPUs or additional accelerators, plus optional Wi-Fi 6E and Bluetooth 5.3 modules.
Display outputs are equally robust: one HDMI 2.1 (4K@60 Hz), one DP 1.4a, and one USB-C (2K@60 Hz) support multi-monitor development environments. Despite this capability, the unit maintains a compact, minimalist industrial design with an efficient thermal structure, keeping typical power draw between 10 W and 75 W—balancing desktop elegance with engineering practicality.
A Complete Software Ecosystem
Hardware innovation alone is not enough; accessible software is equally critical. Nanyang Singtech plans to equip the SingNova-H Studio with a mature, full-stack software ecosystem based on Linux and other open-source operating systems. The platform includes a professional programming toolchain and compatibility with mainstream AI frameworks, lowering the barrier for developers, researchers, and creatives building on open hardware.
Cross-Industry Application Scenarios
The SingNova-H Studio spans multiple verticals. In enterprise settings, it serves as a private AI inference engine for secure deployment of generative models in document analysis, code generation, and knowledge management. For security and smart-city operators, the system supports parallel camera access, delivering real-time video analytics—facial recognition, behavioral analysis, and video structuring—at 128-channel 1080p@30 fps decode or 64-channel encode.
In industrial and IoT environments, it acts as a high-reliability edge-computing node for smart factories, robotics, and autonomous driving, providing the deterministic low latency required for real-time decision-making and closed-loop control. Creative professionals benefit from accelerated Stable Diffusion text-to-image generation, video editing, and 3D creation, while universities and research institutes leverage the open-source RISC-V architecture for ecosystem development, algorithmic research, and experimental teaching.
A Milestone for Singapore’s Deep-Tech Ecosystem
Nanyang Singtech’s debut at ATxSG highlights Singapore’s emergence as an innovation hub for advanced computing hardware. By combining NTU’s research heritage with commercial engineering execution, the company is positioning itself as a key infrastructure provider in the global shift toward localized, efficient, privacy-preserving AI.
"SingNova-H Studio embodies our belief that the future of intelligent computing must be open, local, and scalable," the company stated at the event. "We are not merely building a faster PC; we are redefining what an edge intelligent workstation can deliver for developers and enterprises worldwide."
About Nanyang Singtech
Nanyang Singtech is a Singapore-based deep technology company incubated by Nanyang Technological University. The company 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: 89192557
In case of encountering any inaccuracies, problems, or queries arising from the content shared in this press release that necessitate action, or if you require assistance with a press release takedown, we urge you to notify us at error@releasecontact.com (it is important to note that this email is the authorized channel for such matters, sending multiple emails to multiple addresses does not necessarily help expedite your request). Our responsive team will be readily available to promptly address your concerns within 8 hours, resolving any identified issues diligently or guiding you through the necessary steps for removal. The provision of accurate and dependable information is our primary focus.
