NVIDIA DGX B300

rack
8× B300
NVSwitch + NVLink
2025
FP32
0.7
PFLOPS
TF32
10.0
PFLOPS
Power
14.5
kW Total
Memory
2150
GB Total
FP32
0.67
PFLOPS
TF32
10.00
PFLOPS
FP16
20.00
PFLOPS
BF16
20.00
PFLOPS

System Details

GPU Configuration

GPU Count: 8 GPUs
Architecture: Blackwell Ultra
Interconnect: NVSwitch + NVLink

System Specifications

Form Factor: rack
Total Power: 14.5 kW
Total Memory: 2150 GB
Memory Bandwidth: 64000 GB/s

Precision Performance Breakdown

PrecisionSystem PerformancePer GPUEfficiency
FP32 0.666 PFLOPS 83.3 TFLOPS
TF32 10.000 PFLOPS 1250.0 TFLOPS
FP16 20.000 PFLOPS 2500.0 TFLOPS
BF16 20.000 PFLOPS 2500.0 TFLOPS
FP8 40.000 PFLOPS 5000.0 TFLOPS
FP4 120.000 PFLOPS 15000.0 TFLOPS

Powered by NVIDIA B300

This system utilizes 8 × NVIDIA B300 SXM 288GB GPUs, each delivering exceptional performance for AI and HPC workloads.

Per GPU TDP

1400W

Per GPU Memory

288 GB

Process Node

4nm

Architecture

Blackwell Ultra

Documentation & Resources

workstation-datasheet-dgx-spark-gtc25-spring-nvidia-us-3716899-web.pdf

NVIDIA • 2025-10-16

View Document ↗

NVIDIA DGX Documentation

System guides, deployment resources

Browse ↗

Typical Use Cases

Large Language Model Training
Distributed Deep Learning
Multi-GPU Inference
HPC Simulations
Scientific Computing

The NVIDIA DGX B300 with 8× B300 GPUs is designed for enterprise-scale workloads requiring 0.7 PFLOPS of compute power with NVSwitch + NVLink interconnect for efficient multi-GPU communication.