NVIDIA B100 SXM 192GB
Blackwell SXM 2024 4nm
FP32
60.0
TFLOPS
VRAM
192 GB
TDP
700 W
85.7 TFLOPS/kW
Bandwidth
8.0 TB/s
memory
Performance Metrics
Peak theoretical throughput by precision type
| Precision | Peak TFLOPS | Efficiency |
|---|---|---|
| INT8 | 3500.0 | 5.000 TFLOPS/W |
| FP8 | 3500.0 | 5.000 TFLOPS/W |
| FP4 | 7000.0 | 10.000 TFLOPS/W |
| FP16 | 1750.0 | 2.500 TFLOPS/W |
| BF16 | 1750.0 | 2.500 TFLOPS/W |
| TF32 | 875.0 | 1.250 TFLOPS/W |
| FP32 | 60.0 | 0.086 TFLOPS/W |
| FP64 | 30.0 | 0.043 TFLOPS/W |
FP8 Efficiency
5.000 TFLOPS/W
3500.0 TFLOPS / 700W
FP16 Efficiency
2.500 TFLOPS/W
1750.0 TFLOPS / 700W
FP32 Efficiency
0.086 TFLOPS/W
60.0 TFLOPS / 700W
Power Specifications
TDP
700 W
Max Power
805 W
Power Connector
PCIe 16-pin
Cooling
Liquid
Memory Specifications
Capacity
192 GB
Type
--
Bandwidth
8000 GB/s
Interface
--
Hardware & Design
Form Factor
SXM
Architecture
Blackwell
Process Node
4nm
Launch Year
2024
Variant
SXM 192GB
Market Segment
Data Center
Documentation & Resources
Common Use Cases
General Compute AI/ML Workloads Data Processing
The NVIDIA B100 is optimized for high-performance computing tasks with Blackwell architecture delivering 60 TFLOPS of compute power.
Where to Rent
Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.
Browse GPU Cloud ProvidersSimilar GPUs
NVIDIA RTX PRO 6000 Blackwell Workstation Edition
PCIe · 2025
FP32: 125.0 TFLOPS
Compare vs RTX PRO 6000 BlackwellSystems Using This GPU
Pre-configured systems featuring the NVIDIA B100
| System | Peak Performance |
|---|---|
NVIDIA HGX B100 8-GPU HGX · baseboard · 2024 | 56.00 PFLOPS |
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
Loading verification...
No spam, unsubscribe anytime.