- Advantech unveils cutting-edge Edge AI server solution for generative AI, featuring Phison’s aiDAPTIV+ technology.
- The AIR-520 Edge AI Server integrates AMD EPYC 7003 series processor, SQ ai100 AI SSDs, NVIDIA RTX GPU cards, Edge AI SDK, and NVIDIA AI Enterprise.
- Tailored large language models (LLMs) enhance enterprise knowledge management, automating data organization and boosting productivity.
- Four solution options cater to various applications: L13B for real-time tasks, L33B for complex tasks, L70B for sophisticated data analysis, and L70B-Plus with NVIDIA AI Enterprise for optimized AI SDKs.
- Streamlined installation with SQ ai100 AI SSDs extends GPU vRAM for cost efficiency and compact design.
- Advantech provides Edge AI SDK with GenAI Training Studio and DeviceOn for efficient AI orchestration and maintenance.
- Early-release versions of AIR-520 Edge AI Server Solution are available now.
Main AI News:
Advantech, a prominent figure in the realm of Industrial IoT, is excited to unveil a revolutionary Edge AI server solution tailored for generative AI applications, showcasing Phison’s cutting-edge aiDAPTIV+ technology. The AIR-520 Edge AI Server, fueled by the formidable AMD EPYC 7003 series processor, seamlessly integrates SQ ai100 AI SSDs, NVIDIA RTX GPU cards, an Edge AI SDK, and NVIDIA AI Enterprise, culminating in a plug-and-play solution designed for swift deployment.
Pioneering generative AI tools like large language models (LLMs) are reshaping the landscape of enterprise knowledge management, revolutionizing data organization, retrieval, and analysis, thereby elevating productivity and refining decision-making processes. Tailored LLMs significantly enhance precision, while edge training bolsters data privacy, albeit at a potentially higher cost. This innovative solution facilitates LLM fine-tuning with 1-4 GPU cards and SQ ai100 AI SSDs, empowering enterprises to train LLMs efficiently while safeguarding sensitive data at the edge.
Innovative Advantech AIR-520 Edge AI Server for Generative AI
Diverse Solutions Catering to Various Applications
Advantech presents four distinct options: AIR-520-L13B/L33B/L70B, and L70B-Plus, meticulously crafted to address a spectrum of scales and applications. The L13B variant is ideally suited for real-time applications such as chatbots and language translation. Meanwhile, the L33B variant is engineered for more intricate tasks, augmenting productivity and fostering innovation in content creation. For endeavors requiring sophisticated data analysis and decision-making within specialized domains, the L70B model stands as an unparalleled choice. Additionally, the L70B-Plus, equipped with the NVIDIA AI Enterprise, delivers end-to-end, dependable, and optimized AI SDKs accompanied by extensive long-term support and expert consultation services.
Streamlined Installation Geared Towards Cost Efficiency
All AIR-520 solutions come complete with SQ ai100 AI SSDs, harnessing the prowess of Phison’s aiDAPTIV+ technology. These SSDs seamlessly extend the GPU vRAM, facilitating LLM fine-tuning with minimal GPU cards. This streamlined approach not only surmounts budget constraints but also renders the Edge AI Server more compact compared to conventional large rack-mount servers. With a footprint akin to a desktop PC, it can be effortlessly rack-mounted using compatible accessories, effectively mitigating concerns regarding space constraints and maintenance overheads.
Accelerated Development Through Software Services
In tandem with LLM fine-tuning capabilities, Advantech offers an Edge AI SDK featuring the GenAI Training Studio, preloaded with Llama-2 models tailored for applications such as chatbots and data analysis. This simplifies and expedites the process of training customer-specific LLM models and evaluating inferences on the AIR-520 platform. Moreover, Advantech’s DeviceOn streamlines OTA software/container updates and remote management, fostering seamless AI orchestration and ensuring sustained operational efficiency.
Conclusion:
Advantech’s introduction of the AIR-520 Edge AI Server Solutions signifies a significant advancement in the field of generative AI. With its integration of state-of-the-art technologies and tailored solutions for diverse applications, it sets a new standard for edge computing in the market. This innovation not only enhances productivity and decision-making processes but also underscores the growing demand for efficient and secure AI solutions across various industries.
