Supermicro AS -4125GS-TNHR2-LCC GPU A+ Server DP AMD 4U Liquid-Cooled System with NVIDIA HGX H100/H200 8-GPU
High-performance A+ Server with dual AMD EPYC™ 9004/9005 processors, liquid-cooled design, and support for up to 8 NVIDIA HGX H100/H200 GPUs. Ideal for AI training, deep learning, and high-compute workloads.

Supermicro AS-4125GS-TNHR2-LCC Resources
Engineered for Maximum AI Performance
Supermicro AS-4125GS-TNHR2-LCC is a purpose-built GPU A+ Server designed for demanding AI training, HPC, and data analytics workloads with dual AMD EPYC 9004/9005 processors and support for up to 8 NVIDIA HGX H100/H200 GPUs interconnected via NVIDIA NVLink with NVSwitch™.


Built for Scalability and Reliability
Accelerate compute-intensive AI workloads with 24 DIMM slots supporting up to 6TB DDR5 memory and 8 high-performance NVIDIA GPUs, while benefitting from advanced liquid cooling and redundant Titanium-level power supplies. Supports dual AMD EPYC™ 9004/9005 CPUs up to 400W TDP. Efficient PCIe Gen5 interconnect between CPUs and GPUs for optimal data throughput. Trusted Platform Module (TPM 2.0) and Silicon Root of Trust for enhanced security. Designed for data-intensive applications in AI, scientific computing, and big data

Reliable Performance for Your Enterprise
Ensure uninterrupted performance with high-efficiency cooling and redundant power supplies, capable of operating in demanding data center environments.
- Redundant 4x 5250W Titanium Level (96%) power supplies for maximum uptime
- Advanced thermal management with D2C Cold Plate for efficient heat dissipation
- Support for 8 PCIe Gen5 x16 slots to maximize expandability
- Optimized for AI model training, big data analytics, and virtualization
MFG Number | AS-4125GS-TNHR2-LCC |
---|---|
Condition | Item Condition : Brand New |
Price | $0.00 |
Show Product Will Be Available Soon on FE | No |
Estimated time of Delivery | Estimated Delivery 4 Weeks. |
Product Card Description | Supermicro AS-4125GS-TNHR2-LCC GPU A+ Server features dual AMD EPYC™ 9004/9005 CPUs, 24 DIMM slots (up to 6TB DDR5 memory), and support for up to 8 NVIDIA HGX H100/H200 GPUs interconnected via NVIDIA NVLink with NVSwitch™. Equipped with advanced liquid cooling and redundant Titanium power supplies, designed for AI training and compute-intensive applications. |
Order Processing Guidelines | Order Processing Guidelines:
Inquiry First – Please reach out to our team to discuss your requirements before placing an order. |
Special Price | $0.00 |
Special Price From Date | May 2, 2024 |