Atlas 300I Pro Inference Card
Model:Atlas 300I Pro
Parameters6ItemHigh-performance accelerator card for AI inference. Supports multiple AI frameworks, ideal for video analytics and natural language processing.
- Ascend 310P Low-Power Inference
- 140 TOPS INT8 Compute
- PCIe standard interface compatible
- Broad support for multiple frameworks
Technical Specifications
Atlas 300I Pro Inference CardCommon6Item parameters, by3Displayed by group; actual configuration subject to project selection confirmation.
| Compute & Memory4Item | |
|---|---|
| AI Chip | Ascend 310P |
| INT8 Compute | 140 TOPS |
| FP16 Compute Power | 70 TFLOPS |
| VRAM | 2 4GB LPDDR4X |
| Interfaces and Extensions1Item | |
| API | PCIe 4.0 x16 |
| Power and Cooling1Item | |
| Power Consumption | 75W |
Similar Models
TaiShan 200 AI Server 2280A
Model:TaiShan 200 2280A
2U rack server based on Kunpeng 920 processor: high performance, high reliability, and high security for enterprise core workloads, virtualization, big data, and more.
View DetailsTaiShan 200 High-Density Server 2480
Model:TaiShan 200 2480
4U high-density server based on Kunpeng 920 processors, supporting quad-processor configurations. Delivers ultra-high compute density for large-scale databases, ERP systems, and mission-critical applications.
View DetailsTaiShan 100 Server 2280
Model:TaiShan 100 2280
2U entry-level rack server based on Kunpeng 920 processor; high cost-performance ratio, ideal for web services, distributed storage, and lightweight computing.
View DetailsInquireAtlas 300I Pro Inference Card
Leave your configuration requirements and quantity, and we'll provide selection recommendations and quotes tailored to your project scenario.
Purchase Intent
Submit your requirements and contact details. A sales consultant will confirm pricing, stock availability, and delivery options with you. Product prices are subject to quotation.