
TITAN GM645R-G6
AI & GPU Server
- 4U NVIDIA MGX™
- Dual AMD EPYC™ 9005 Series processors
- Supports 8 GPUs with 600W TDP each
- Up to 4 TB DDR5 6400 MT/s memory
- 8 x E1.S NVMe drive bays and 2 x M.2
- ASPEED AST2600 management
- 3200W (3+1) 80 PLUS Titanium Power Supply
- Perfect for Generative AI, Agentic AI, Industrial & Physical AI, Data Analytics, Simulation, and AI-Accelerated Rendering

Product Overview
The CIARA TITAN GM645R-G6 is a 4U NVIDIA MGX™ server built for Generative AI, Agentic AI, Industrial & Physical AI, Data Analytics, Simulation, and AI-Accelerated Rendering. Powered by dual AMD EPYC™ 9005 Series processors (up to 400W TDP each), it delivers robust multi-core compute performance for data-intensive and GPU-accelerated workloads.
Designed on the NVIDIA MGX™ modular architecture, the system supports up to eight NVIDIA GPUs, including the NVIDIA RTX PRO™ 6000 Blackwell Server Edition, H200 NVL, H100 NVL, and L40S, providing optimized airflow, slot spacing, and power delivery for sustained GPU performance under continuous load.
With support for up to 4 TB of DDR5 6400 MT/s RDIMM/3DS RDIMM memory, eight E1.S NVMe drive bays, dual M.2 storage devices, and up to eight 400Gb/s QSFP112 ports via NVIDIA ConnectX®-8, the CIARA TITAN GM645R-G6 delivers the networking throughput, memory bandwidth, storage density, and scalability required for modern AI clusters and next-generation high-power GPU deployments.


Key Features and Benefits
AMD EPYC™ Compute for Advanced AI Workloads
The CIARA TITAN GM645R-G6 features dual AMD EPYC™ 9005 Series processors (up to 400W TDP each), delivering powerful multi-core compute performance for Generative AI, Agentic AI, Industrial & Physical AI, Data Analytics, Scientific Simulation, and AI-Accelerated Rendering. This architecture efficiently processes data and sustains high GPU utilization during intensive multi-node operations.
MGX™ Architecture for GPU-Optimized Performance
Built on NVIDIA MGX™ modular platform, the system supports up to eight NVIDIA GPUs (600W TDP each), including NVIDIA RTX PRO™ 6000 Blackwell Server Edition, H200 NVL, H100 NVL, and L40s, with standardized airflow, power delivery, and slot spacing. This ensures consistent cooling, stable multi-GPU performance, and flexibility to adopt next-generation GPUs.
High-Speed Networking for AI Clusters
The system includes eight 400Gb/s QSFP112 ports powered by NVIDIA ConnectX®-8, delivering ultra-high network bandwidth for distributed training, multi-node inference, and accelerated data movement across AI environments. This level of connectivity enables efficient scaling for modern high-performance AI clusters.
High-Bandwidth Memory and E1.S Storage Architecture
With support for up to 4 TB of DDR5 6400 MT/s RDIMM/3DS RDIMM memory, eight E1.S NVMe drive bays, dual M.2 storage devices, and up to eight 400Gb/s QSFP112 ports via NVIDIA ConnectX®-8, the CIARA TITAN GM645R-G6 delivers the networking throughput, memory bandwidth, storage density, and scalability required for modern AI clusters and next-generation high-power GPU deployments.
Perfect For
Gen AI
Agentic AI
AI-Accelerated Rendering
Industrial AI
Physical AI
Data Analytics
Simulation
Tech Specs
Resources
Product Downloads
Service & Support
Related Products













