MSI CG480-S6053 AI Server
CG480-S6053Available to order6 week lead time
Built on MSI's CG480-S6053 — a 4U dual-socket platform derived from the NVIDIA MGX reference design — this system supports dual AMD EPYC 9005 series processors and up to eight double-wide PCIe 5.0 GPUs. Choose between two validated GPU configurations below: eight NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs paired with dual EPYC 9555 processors, or eight NVIDIA H200 NVL GPUs (NVLink-bridged) paired with dual EPYC 9575F processors for maximum GPU-to-GPU bandwidth. Every unit ships fully assembled through L6-to-L10 system integration and test, backed by a 3-year warranty.
- CPU
- 2x AMD EPYC 9555 (Zen 5 “Turin”), 64 cores / 128 threads each — 128 cores / 256 threads total
- Clock Speed
- 3.2GHz base, up to 4.4GHz max boost
- Cache
- 256MB L3 per CPU (512MB total)
- TDP
- 360W default per socket (320W–400W configurable)
- CPU
- 2x AMD EPYC 9575F (Zen 5 “Turin”), 64 cores / 128 threads each — 128 cores / 256 threads total
- Clock Speed
- 3.3GHz base, up to 5.0GHz max boost (4.5GHz all-core boost)
- Cache
- 256MB L3 per processor (512MB total), 64MB L2 per processor
- TDP
- 400W per processor (configurable 320W–400W)
- GPU
- 8x NVIDIA RTX PRO 6000 Blackwell Server Edition
- GPU Memory
- 96GB GDDR7 with ECC per GPU — 768GB aggregate
- Memory Bandwidth
- up to 1,597 GB/s per GPU
- Compute Cores
- 24,064 CUDA cores, 752 5th-gen Tensor cores, 188 4th-gen RT cores per GPU
- FP32 Performance
- 120 TFLOPS peak per GPU
- GPU Interconnect
- PCIe Gen5 x16 host interface per card, 128GB/s bidirectional — NVLink not supported on this SKU
- GPU Power
- up to 600W configurable per card, passively cooled
- GPU
- 8x NVIDIA H200 NVL (Hopper architecture), PCIe Gen5, dual-slot, air-cooled
- GPU Memory
- 141GB HBM3e per GPU (1,128GB total), 4.8TB/s memory bandwidth per GPU
- GPU TDP
- Up to 600W per GPU (configurable)
- Tensor Performance
- 3,958 TFLOPS FP8 / 1,979 TFLOPS FP16-BF16 (with sparsity) per GPU
- Host Interface
- PCIe Gen5 x16 per GPU (128GB/s)
- NVLink Bridges
- 2x 4th-generation NVLink bridge (2-/4-way bridge configuration)
- Bridged Bandwidth
- Up to 900GB/s GPU-to-GPU bandwidth within bridged groups
- System Memory
- 768GB DDR5 RDIMM ECC (12x 64GB modules, 12 of 24 DIMM slots populated)
- Speed
- up to 6400 MT/s
- System Memory
- 1,536GB DDR5 RDIMM ECC (24x 64GB modules, all 24 DIMM slots populated)
- Speed
- up to 6400 MT/s
- Boot Drive
- 1x 960GB M.2 NVMe SSD (M.2 2280/22110, PCIe 3.0 x2 chassis slot)
- Data Storage
- 2x 3.84TB U.2 NVMe SSD (7.68TB total), front hot-swap 2.5" SFF bay
- Expansion
- Chassis supports up to 8–12x front hot-swap 2.5" SFF bays (U.2 PCIe 5.0 NVMe / SAS / SATA) for additional storage
- Onboard LAN
- 2x 10GBase-T RJ-45 (Intel X710-AT2)
- Management LAN
- 1x dedicated 1000Base-T BMC port
- Power Supplies
- 4x hot-swap redundant, 3200W CRPS, 80 PLUS Titanium (3+1 redundant configuration)
- Form Factor
- 4U rackmount, NVIDIA MGX reference design (MSI CG480-S6053)
- Dimensions
- ~800mm (L) x 440mm (W) x 175mm (H)
- Weight
- ~35kg (bare chassis, excludes installed components)
- GPU Bay Support
- Up to 8x FHFL dual-width PCIe 5.0 GPUs, air-cooled
- BMC
- ASPEED AST2600, IPMI 2.0, DMTF Redfish
- Security
- TPM 2.0, dual BIOS
- Certification
- Platform listed in Red Hat's hardware catalog for RHEL 9.5+ and RHEL AI 1.5+ (x86_64)
- Integration
- Full L6-L10 system assembly, burn-in, and validation testing
- Warranty
- 3-year system warranty
Processor & Chipset — RTX PRO 6000 Configuration
Processor & Chipset — H200 NVL Configuration
GPU — RTX PRO 6000 Configuration
GPU — H200 NVL Configuration
Memory — RTX PRO 6000 Configuration
Memory — H200 NVL Configuration
Storage
Networking
Power
Chassis
Management & Security
Assembly & Support
Priceavailable on request
Get a quote