Product Features

MXFP4 Low-precision Inference
  • Achieve industry-leading compute performance of 12.4 PFLOPS @ mxFP4
  • Support HiF8 / mxFP8 / mxFP4 data formats,significantly improving quantization acceleration performance
Ultra-high Memory Bandwidth
  • Deliver industry-leading memory R/W bandwidth of 4.0 TB/s
  • 768 GB on-chip memory per server
Ultra-high UB Link Bandwidth
  • Provide 13.4 TB/s UB Link bandwidth per server (16-NPU FM)
  • Support 16-NPU UB full-mesh interconnect across 2 servers

Atlas 650E Server Product Specifications

Form Factor

14U AI server

NPU

8 × 950DT NPU

CPU

2 × 950 CPU (up to 96 cores, 2.3GHz)

*AI Performance

12.48 PFLOPS mxFP4

6.43 PFLOPS mxFP8/FP8/HiF8

3.40 PFLOPS FP16/BF16

On-chip Memory

8 × 96 GB total, 4.0 TB/s peak memory bandwidth

System Memory

Support 24 memory slots running at up to 6,400 MT/s

Support up to 64 GB per DDR5 RDIMM

Storage

8 x 2.5 NVMe + 2 x 2.5 SATA (with RAID card configured)

UB Link Bandwidth

8-NPU Server: 8 x 784 GB/s (bidirectional)

16-NPU Full-mesh: 2 x 8 x 1.68 TB/s (bidirectional)

Networking

UBoE: Providing 400 Gbps per NPU (unidirectional)

RoCE: Providing 400 Gbps per NPU (unidirectional)

Bus Type

PCIe 5.0 x 5

Power Modules

6 x 3.0 kW power supply modules, supporting hot swapping & 5+1 redundancy

Power Supply

220VAC

336HVDC/240HVDC

Cooling

Air cooling

Fan Modules

30 fan modules, supporting hot swapping & N+1 redundancy

Power Usage

~14.5 kW

Operating Temperature

5℃~35℃(41℉~95℉)

Weight

251 kg

Dimensions

618.8 mm (Height) x 447mm (Width) x 920mm (Length)

Application Scenarios

For more industry solutions, check "Solutions-Industry Applications".

AI Agent

Financial Risk Control

Office Assistant

Content Review

AI Coding

AI Customer Service