Lenovo WenTian WA5480 G3 С500 512GB
The Lenovo WenTian WA5480 G3 is a next-generation 4U dual-socket AI training server engineered for extreme flexibility and performance.
Supporting 4th and 5th Gen Intel® Xeon® Scalable processors and the latest PCIe 5.0 standard, it can accommodate up to 8 multi-type, multi-brand AI accelerator cards like MetaX, Moore Threads, Iluvatar.
Its unique hardware design allows for multiple CPU-GPU topologies - Direct Connect, Balance, and Common - enabling users to optimize the system for specific AI workloads.
With N+N redundant power supplies and comprehensive functional monitoring of key components, the WA5480 G3 provides a secure and stable environment for mission-critical AI operations.
CPUs
2x Xeon Gold 6430 (64 ядра / 128 потоков)
RAM
1ТБ (16x64 ГБ DDR5-4800MT/c ECC RDIMM)
Accelerators
8x MetaX C500 64GB
Storage
2x1.92ТБ NVMe U.2 SSD + 2x 480 ГБ SATA SSD + SATA SW RAID
Pre-order
Product status
CPU & Memory
CPU
2x Xeon Gold 6430 (64 ядра / 128 потоков)
RAM
1ТБ (16x64 ГБ DDR5-4800MT/c ECC RDIMM)
RAM Type
DDR5
Storage Configuration
SSD Model
-
Total Storage
2x1.92ТБ NVMe U.2 SSD + 2x 480 ГБ SATA SSD + SATA SW RAID
Disk Interface
-
GPU Configuration
GPU Vendor
MetaX Integrated Circuits
GPU Model
MetaX C500 64GB
GPU Count
8
Total VRAM
512ГБ HBM2E
Interconnect
MetaXLink
Network & Virtualization
Network Interfaces
1x 25 Гбит/с SFP28 OCP NIC + 1x 400 Гбит/с OSFP NDR IB
Virtualization (MIG)
Поддерживается (vGPU на базе SR-IOV)
CUDA Compatibility
MXMACA
Power & Thermal
Power Supply
4x 2700Вт Platinum
TDP
5500
Cooling Type
Пассивное
Server Specifications
Technical ID
Lenovo WenTian WA5480 G3 С500
Platform
Lenovo WenTian WA5480
Form Factor
4U
Warranty
3
Primary Tasks
ИИ-вычисления, анализ данных, обучение нейросетей
Price
On request