Aiserver Manufacturer
1 / 5
| Integrated GPU | The Si N260 (MXN260) 64GB is a domestic high-end training and inference integrated GPU, optimized for AI large model training and scientific computing. |
| Performance Tier | Comparable to L20 |
| Compute Power |
FP32: 15 TFLOPS FP16: 80 TFLOPS INT8: 160 TOPS |
| Memory Architecture | HBM2E High Bandwidth Memory, offering ~30% higher bandwidth than HBM2, ideal for large model inference. |
| Process Technology | TSMC 7nm FinFET process with 2.5D CoWoS advanced packaging for heat dissipation and stability. |
| Memory Bandwidth | 1.2TB/s Bandwidth with 1024-bit Bus Width and 9.375Gbps effective frequency. |
| Connectivity | PCIe Gen4.0 x16 (64GB/s bidirectional) supporting x86/ARM host interconnection. |
| Form Factor | FHFL Dual-Slot, Weight: 1064g. |
Security is integrated into every phase of the lifecycle, including protected supply chain and factory-to-site integrity assurance. Silicon-based root of trust anchors end-to-end boot resilience.
Increase efficiency and accelerate operations with comprehensive systems management solutions. Simplify, automate and centralize management with advanced enterprise consoles.
Designed with recycled materials and innovative options for energy efficiency, helping to reduce carbon footprint and lower operation costs throughout the product lifecycle.
Designed for 24x7 enterprise data center operations, featuring components optimized for deployment at scale. Supports Secure Boot with Hardware Root of Trust technology to meet new data center standards.
The passively cooled, energy-efficient design fits a wide range of Certified systems and is easy to integrate into existing data center infrastructure.
