ASUS Announces ESC A8A-E12U Support for AMD Instinct MI350 Series GPUs
Enhanced performance and seamless upgrade path
for next-generation AI and HPC deployments
- Seamless upgrades: ESC A8A-E12U now supports AMD Instinct™ MI350 Series GPUs for next-gen AI and HPC performance
- Maximum throughput: 288GB HBM3E and 8TB/s bandwidth per GPU accelerate large AI and scientific workloads
- Secure efficiency: Built-in security and reduced GPU count lower cost and complexity for multi-tenant data centers
TAIPEI, Taiwan, May 19, 2025 — ASUS today announced that its flagship high-density AI server, ESC A8A-E12U, now supports the latest AMD Instinct™ MI350 series GPUs. This enhancement empowers enterprises, research institutions, and cloud providers to accelerate their AI and HPC workloads with next-generation performance and efficiency — while preserving compatibility with existing infrastructure.
Built on the 4th Gen AMD CDNA™ architecture, AMD Instinct MI350 series GPUs deliver powerful new capabilities, including 288GB of HBM3E memory and up to 8TB/s of bandwidth — enabling faster, more energy-efficient execution of large AI models and complex simulations. With expanded support for low-precision compute formats such as FP4 and FP6, the Instinct MI350 series significantly accelerates generative AI, inference, and machine-learning workloads. Importantly, Instinct MI350 series GPUs maintain drop-in compatibility with existing AMD Instinct MI300 series-based systems, such as those running Instinct MI325X — offering customers a cost-effective and seamless upgrade path. These innovations reduce server resource requirements and simplify scaling and workload management, making Instinct MI350 series GPUs an ideal choice for efficient, large-scale AI deployments.
ASUS ESC A8A-E12U with AMD Instinct MI350X delivers scalable, high-performance AI and HPC
The newly-enhanced ASUS ESC A8A-E12U is a high-performance 7U server built to meet the growing demands of AI and HPC workloads. Paired with dual AMD EPYC™ 9005 processors, the ESC A8A-E12U server leverages the MI350X’s architecture to deliver outstanding performance for a range of demanding scenarios, including training and inferencing large language models (LLMs), fine-tuning generative AI, and running scientific computing simulations. With 288GB of high-bandwidth memory per GPU and up to 8TB/s bandwidth, fewer GPUs are needed to support larger models—reducing server density and simplifying infrastructure planning. This allows organizations to scale more efficiently, minimize power consumption, and reduce total cost of ownership in AI-driven data centers.
In addition, the MI350X platform integrates advanced security features such as Secure Boot, DICE attestation, SR-IOV for secure multi-tenant virtualization, and GPU-to-GPU communication encryption, making ESC A8A-E12U an ideal solution for multi-tenant environments where data confidentiality and regulatory compliance are critical — such as in government, finance, and healthcare sectors.
“As AI workloads continue to evolve in complexity and scale, ASUS is committed to delivering infrastructure solutions that help our customers move faster,” said Paul Ju, General Manager of the ASUS Server Business Unit. “With support for AMD Instinct MI350X GPUs, our ESC A8A-E12U server provides a robust, future-ready platform for next-era AI and HPC deployments.”
ESC A8A-E12U continues ASUS leadership in enterprise and cloud infrastructure, providing scalable and reliable platforms that support the latest in compute acceleration technologies. ASUS remains committed to delivering innovations that unlock new possibilities in AI, HPC, and advanced data center workloads.
