Skip to Content Facebook Feature Image

ASRock Industrial Launches AI BOX-A395, Compact AI Workstation Powered by AMD Ryzen™ AI Max

Business

ASRock Industrial Launches AI BOX-A395, Compact AI Workstation Powered by AMD Ryzen™ AI Max
Business

Business

ASRock Industrial Launches AI BOX-A395, Compact AI Workstation Powered by AMD Ryzen™ AI Max

2026-03-23 09:00 Last Updated At:09:15

TAIPEI, March 23, 2026 /PRNewswire/ -- As the era of pervasive AI reshapes industries worldwide, ASRock Industrial today announced the AI BOX-A395, a compact yet powerful AI box that brings the performance of an ultimate AI workstation into a single system. Powered by AMD Ryzen™ AI Max+ 395 processors, delivering up to 50 TOPS of AI acceleration while integrating CPU, GPU, and NPU within a compact system. With support for up to 128GB LPDDR5x-8000 unified memory, it enables large AI models and data-intensive workloads to run directly on-device, delivering responsive AI processing while reducing reliance on cloud infrastructure. Designed for enterprises, developers, and system integrators, the AI BOX-A395 supports the AI everywhere ecosystem by translating large-scale AI capabilities into practical local deployment. By combining high compute density, integrated AI acceleration, and rich I/O connectivity, the system provides a scalable foundation for applications ranging from AI model and application development, engineering and 3D design, and high-resolution content creation and media production.

  • AMD Ryzen™ AI Max+ 395 Processors – The Most Powerful x86 APU

The AI BOX-A395 features AMD Ryzen™ AI Max+ 395 processors with Zen 5 CPUs of up to 16 cores and 32 threads, AMD Radeon™ 8060S Series GPU, and AMD XDNA™ 2 NPU delivering up to 50 TOPS AI acceleration. Supporting both Windows and Linux OS, the system brings ultimate AI workstation performance in a compact form factor, enabling advanced local AI inference, content creation, and AI development workloads.

  • Up to 128GB LPDDR5x Unified Memory Power Your AI Performance

Supporting up to 128GB LPDDR5x-8000 unified memory, the system is engineered to handle large AI models and memory-intensive workloads efficiently on-device, enabling smooth AI inference and fast data processing without reliance on external accelerators.

  • Run Large AI Models at the Edge

Bring AI closer to where data lives. AI BOX-A395 enables large language models, generative AI, and vision AI to run locally, delivering instant AI responses, lower cloud costs, and full control oversensitive data.

  • 10GbE and USB4 Power High-Speed Edge AI

The platform offers comprehensive I/O, including two USB4, one USB 3.2 Gen2 Type-C, two USB 3.2 Gen2, and two USB 2.0 ports, along with dual Ethernet featuring one 10GbE and one 2.5GbE LAN. This high-bandwidth connectivity enables fast data transfer for demanding local AI workloads. The system also supports quad-display output through two HDMI 2.1 and three DisplayPort 2.1 interfaces, delivering resolutions up to 8K.

  • Dual Storage and Wi-Fi 7 for High-Speed AI Workloads

The platform provides two M.2 Key M slots (2242/2280) with PCIe Gen4 x4 support for SSDs and RAID 0/1, along with one M.2 Key E slot supporting Wi-Fi 7 and Bluetooth 5.4. This configuration ensures scalable storage performance and flexible wireless expansion for diverse deployment scenarios.

  • Advanced Thermal Architecture for Sustained AI Performance

AI BOX-A395 features 6 heat pipes thermal architecture with a copper base and optimized airflow design to efficiently dissipate heat from high-performance AI workloads. This advanced cooling system enables sustained AI processing while maintaining low acoustic levels, delivering stable and quiet operation for demanding AI development and content creation environments.

  • Enterprise-ready Reliability

Housed in a compact 200 x 100 x 232mm (L x W x H) aluminum chassis with a handle accessory for easy portability, the system includes onboard TPM 2.0, redundant BIOS support, and enterprise-ready reliability and security features to support long-term deployment requirements.

For further information on the AI BOX-A395, please visit our website or reach out via product Inquiry.

About ASRock Industrial

ASRock Industrial Computer Corporation, established as an independent entity in July 2018, is a prominent industry player specializing in the development of motherboards, edge computers, and related solutions for the manufacturing, business, and retail sectors. As the world-leading provider of Industrial PC systems and motherboards, our customers base spans the globe. Previously, it had been a business unit of ASRock Inc. (est. 2002) which was set up in 2011. ASRock Industrial now operates autonomously, allowing us to devote all resources to delivering exceptional B2B solutions.

Our vision is to cocreate an intelligent world, aligning with ASRock Industrial's core dedication to CARES (Commerce/Automation/Robot/Entertainment/Security) industries. With a central R&D design team comprising nearly 58% of our total staff, ASRock Industrial has the resources and expertise to develop cutting-edge, reliable products tailored to meet your business requirements. Our product portfolio encompasses both off-the-shelf offerings and customized solutions catered to the specific needs of OEMs/ODMs.

** This press release is distributed by PR Newswire through automated distribution system, for which the client assumes full responsibility. **

ASRock Industrial Launches AI BOX-A395, Compact AI Workstation Powered by AMD Ryzen™ AI Max

ASRock Industrial Launches AI BOX-A395, Compact AI Workstation Powered by AMD Ryzen™ AI Max

  • Collaborative research on AI-based diagnosis of multiple sclerosis, Building a clinical trial dataset
  • Leveraging large-scale, multi-center brain MRI datasets to detect key lesion biomarkers

SEOUL, South Korea, May 26, 2026 /PRNewswire/ -- Neurophet (Co-CEOs Jake Junkil Been and Donghyeon Kim), an artificial intelligence (AI) solution company for brain disorders diagnosis and treatment, announced today that it will conduct a joint research collaboration with the Vall d'Hebron Institut de Recerca (VHIR), the research institute of Vall d'Hebron University Hospital in Spain. The collaboration will focus on Neurophet AQUA MS, its quantitative imaging software for multiple sclerosis.

The joint study aims to develop and validate AI algorithms that detect multiple sclerosis lesions based on magnetic resonance imaging (MRI). To this end, both parties will secure high-quality, large-scale datasets related to multiple sclerosis and will accelerate the development of new technologies to be integrated into Neurophet AQUA MS.

Neurophet AQUA MS is a software module integrated into Neurophet AQUA, Neurophet's brain imaging analysis solution, and is designed to analyze brain images of patients with multiple sclerosis, a chronic and debilitating disease of the central nervous system. The product received FDA 510(k) Clearance in October 2024.

Multiple sclerosis is a chronic disease affecting the central nervous system, which comprises the brain, spinal cord, and optic nerves. It primarily occurs in young adults aged 20 to 40 and shows a higher prevalence in women. It is also more common among Caucasian populations in North America and Europe, while relatively less prevalent in Asian and Black populations. Considering these regional characteristics, Neurophet AQUA MS is expected to establish itself as a strategic product in the U.S. and European markets, where the prevalence of multiple sclerosis is high.

Under the collaboration, VHIR will provide de-identified brain MRI datasets from clinically diagnosed multiple sclerosis patients, along with expert-annotated ground truth lesion labels. Based on these datasets, Neurophet will develop and validate AI algorithms. In particular, the collaboration will focus on developing detection algorithms for the Central Vein Sign (CVS) and Paramagnetic Rim Lesions (PRL)—key imaging biomarkers emphasized in the latest diagnostic criteria—leveraging a large-scale dataset of over 1,000 patients.

Located in Barcelona, Spain, Vall d'Hebron University Hospital is the largest medical institution in Spain, integrating adult, pediatric, women's, and rehabilitation hospitals. It also played a leading role in establishing the 2024 Revised McDonald Criteria, the international standard for the diagnosis of multiple sclerosis.

Dr. Alex Rovira Cañellas, principal investigator of the Neuroradiology Research Group at VHIR, the hospital's research institute, is widely recognized as a world-leading neuroradiologist in the field of multiple sclerosis. Through this collaboration, Neurophet expects to leverage VHIR's world-class clinical data and expertise to further strengthen the clinical reliability of its AI technology.

Donghyeon Kim, Co-CEO of Neurophet, stated, "This collaboration with VHIR enables us to directly develop and validate Neurophet's AI-based multiple sclerosis analysis technology through a leading institution in Europe. Going forward, we plan to further strengthen the clinical reliability of our AI algorithms based on large-scale datasets."

About Neurophet

Neurophet specializes in developing solutions for diagnosis support, treatment guides, and treatment devices targeting brain disorders based on cutting-edge artificial intelligence (AI) technology. The company was founded in 2016 by Co-CEOs Jake Junkil Been and Donghyeon Kim, who developed the next-generation neuro-navigation system.

Major products include brain MRI analysis software "Neurophet AQUA", PET Image Quantitative Analysis Software "Neurophet SCALE PET", Brain imaging treatment planning software for electric and magnetic brain stimulation "Neurophet tES/TMS LAB", Alzheimer's Disease treatment prescription and monitoring software "Neurophet AQUA AD" for tracking treatment efficacy and side effects, and Multiple Sclerosis image analysis software "Neurophet AQUA MS".

Neurophet has set its top priority to helping patients suffering from brain disorders. Based on expertise in neuroscience, Neurophet will continue to challenge and grow to explore the human brain's health and pioneer solutions for brain diseases with AI technology.

  • Collaborative research on AI-based diagnosis of multiple sclerosis, Building a clinical trial dataset
  • Leveraging large-scale, multi-center brain MRI datasets to detect key lesion biomarkers

SEOUL, South Korea, May 26, 2026 /PRNewswire/ -- Neurophet (Co-CEOs Jake Junkil Been and Donghyeon Kim), an artificial intelligence (AI) solution company for brain disorders diagnosis and treatment, announced today that it will conduct a joint research collaboration with the Vall d'Hebron Institut de Recerca (VHIR), the research institute of Vall d'Hebron University Hospital in Spain. The collaboration will focus on Neurophet AQUA MS, its quantitative imaging software for multiple sclerosis.

The joint study aims to develop and validate AI algorithms that detect multiple sclerosis lesions based on magnetic resonance imaging (MRI). To this end, both parties will secure high-quality, large-scale datasets related to multiple sclerosis and will accelerate the development of new technologies to be integrated into Neurophet AQUA MS.

Neurophet AQUA MS is a software module integrated into Neurophet AQUA, Neurophet's brain imaging analysis solution, and is designed to analyze brain images of patients with multiple sclerosis, a chronic and debilitating disease of the central nervous system. The product received FDA 510(k) Clearance in October 2024.

Multiple sclerosis is a chronic disease affecting the central nervous system, which comprises the brain, spinal cord, and optic nerves. It primarily occurs in young adults aged 20 to 40 and shows a higher prevalence in women. It is also more common among Caucasian populations in North America and Europe, while relatively less prevalent in Asian and Black populations. Considering these regional characteristics, Neurophet AQUA MS is expected to establish itself as a strategic product in the U.S. and European markets, where the prevalence of multiple sclerosis is high.

Under the collaboration, VHIR will provide de-identified brain MRI datasets from clinically diagnosed multiple sclerosis patients, along with expert-annotated ground truth lesion labels. Based on these datasets, Neurophet will develop and validate AI algorithms. In particular, the collaboration will focus on developing detection algorithms for the Central Vein Sign (CVS) and Paramagnetic Rim Lesions (PRL)—key imaging biomarkers emphasized in the latest diagnostic criteria—leveraging a large-scale dataset of over 1,000 patients.

Located in Barcelona, Spain, Vall d'Hebron University Hospital is the largest medical institution in Spain, integrating adult, pediatric, women's, and rehabilitation hospitals. It also played a leading role in establishing the 2024 Revised McDonald Criteria, the international standard for the diagnosis of multiple sclerosis.

Dr. Alex Rovira Cañellas, principal investigator of the Neuroradiology Research Group at VHIR, the hospital's research institute, is widely recognized as a world-leading neuroradiologist in the field of multiple sclerosis. Through this collaboration, Neurophet expects to leverage VHIR's world-class clinical data and expertise to further strengthen the clinical reliability of its AI technology.

Donghyeon Kim, Co-CEO of Neurophet, stated, "This collaboration with VHIR enables us to directly develop and validate Neurophet's AI-based multiple sclerosis analysis technology through a leading institution in Europe. Going forward, we plan to further strengthen the clinical reliability of our AI algorithms based on large-scale datasets."

About Neurophet

Neurophet specializes in developing solutions for diagnosis support, treatment guides, and treatment devices targeting brain disorders based on cutting-edge artificial intelligence (AI) technology. The company was founded in 2016 by Co-CEOs Jake Junkil Been and Donghyeon Kim, who developed the next-generation neuro-navigation system.

Major products include brain MRI analysis software "Neurophet AQUA", PET Image Quantitative Analysis Software "Neurophet SCALE PET", Brain imaging treatment planning software for electric and magnetic brain stimulation "Neurophet tES/TMS LAB", Alzheimer's Disease treatment prescription and monitoring software "Neurophet AQUA AD" for tracking treatment efficacy and side effects, and Multiple Sclerosis image analysis software "Neurophet AQUA MS".

Neurophet has set its top priority to helping patients suffering from brain disorders. Based on expertise in neuroscience, Neurophet will continue to challenge and grow to explore the human brain's health and pioneer solutions for brain diseases with AI technology.

** This press release is distributed by PR Newswire through automated distribution system, for which the client assumes full responsibility. **

Neurophet Launches Joint Study with VHIR on 'Neurophet AQUA MS', Accelerating AI Algorithm Validation

Neurophet Launches Joint Study with VHIR on 'Neurophet AQUA MS', Accelerating AI Algorithm Validation

Recommended Articles