NVIDIAGPU
NVIDIA GB200
Blackwell architecture · 192GB memory · 1800 FP16 TFLOPS · 1000W TDP
Cloud Pricing Today
About the NVIDIA GB200
The NVIDIA GB200 is a high-end Blackwell-generation GPU built for large-scale AI training. With 192GB of HBM3e memory and 1,800 FP16 TFLOPS, it is widely deployed in cloud data centres for training frontier AI models.
Memory (VRAM)
192 GB
FP16 Performance
1800 TFLOPS
Power (TDP)
1000W
Architecture
Blackwell
Common Use Cases
Large-scale AI trainingFoundation model fine-tuningHigh-performance computingDistributed training
Key Facts
- Manufacturer
- NVIDIA
- Architecture
- Blackwell
- Accelerator Type
- GPU
- Primary Use
- training
- Memory (VRAM)
- 192 GB
- FP16 Performance
- 1800 TFLOPS
- Thermal Design Power
- 1000W
Related Accelerators
NVIDIA
NVIDIA GB300
Blackwell · 288GB · 2250 TFLOPS
training
NVIDIA
NVIDIA B200
Blackwell · 192GB · 1800 TFLOPS
training
NVIDIA
NVIDIA B300
Blackwell · 288GB · 2250 TFLOPS
training
NVIDIA
NVIDIA H200
Hopper · 141GB · 990 TFLOPS
training
NVIDIA
NVIDIA H100
Hopper · 80GB · 990 TFLOPS
training
NVIDIA
NVIDIA A100 80GB
Ampere · 80GB · 312 TFLOPS
training
Investment Tool
Calculate NVIDIA GB200 ROI
Estimate payback period, annual returns, and 3-year ROI with live Signwl pricing data.
Track NVIDIA GB200 pricing over time
Get access to historical pricing data, regional analysis, and custom alerts.