
TRIDENT iG610R-G6
Immersion GPU Server
- 1U
- Dual 4th and 5th Gen Intel® Xeon® Scalable Processors
- 3 x PCIe 5.0 x16 FHFLDW and 2 x PCIe 5.0 x16 HHHLSW
- Up to 8TB DDR5-5600 MT/s memory
- 2 x M.2 (PCIe 4.0 x4)
- ASPEED AST2600 management
- 3000W (1+1) Redundant 80 PLUS Titanium Power Supply
- Perfect for AI, HPC, Edge Computing and more

Product Overview
The CIARA TRIDENT iG610R-G6 is an immersion-born GPU server, purpose-built for organizations with high demands in AI and compute-intensive applications. With a streamlined 1U form factor, it integrates three GPUs in a PCIe-based architecture and dual 4th or 5th Gen Intel® Xeon® Scalable processors, providing outstanding processing capability for data-intensive tasks.
Designed with a tiered thermal management system, the CIARA TRIDENT iG610R-G6 maintains low consistent temperatures across critical components, ensuring peak performance and preventing thermal imbalances.
Ideal for Gen AI, Machine Learning, Deep Learning, Inferencing, Fine Tuning, Training Language Models, HPC, Data Analytics, Edge Computing, Research etc.—this server is born for immersion-cooled environments, offering these industries the reliability, scalability, and efficiency essential for smooth support of advanced applications.


Key Features and Benefits
Revolutionary GPU Density
Achieving unprecedented density with up to 150 GPUs per immersion tank these servers outpace traditional AI infrastructure, which typically supports a maximum of only 48 GPUs per rack, enabling organizations to pack significantly more computational power into smaller spaces.
Accelerated Data Processing
Equipped with up to 8TB of memory and enhanced RDMA capabilities (Remote Direct Memory Access) this server accelerates data processing speeds, allowing for smoother, faster AI model training and inference.
Future-ready High-Speed Networking
With 3 x PCIe x16 gen 5 dual width slots for GPU and for networking we have 2 x low-profile slots with 2 x OCP 3.0 this server supports 200Gb/s, 400Gb/s, and soon, 800Gb/s network cards, preparing organizations for the next era of high-speed GPU and data communication and reducing bottlenecks in data-intensive AI applications.
GPU Optimization with Flexible Single or Dual Root Complex
Built to optimize GPU resource utilization, the server offers a single or dual root complex option providing the flexibility to choose the optimal setup for specific AI, ML, or DL workloads.
Perfect For
AI
AI Inference
AI Training
HPC
Data Analytics
Edge Computing
Research
Tech Specs
Resources
Product Downloads
Service & Support
Related Products













