NVIDIA A100 PCIe 40GB
Official
Ampere PCIe 2020 7nm
FP32
19.5
TFLOPS
VRAM
40 GB
TDP
250 W
78.0 TFLOPS/kW
Bandwidth
1.6 TB/s
memory
Performance Metrics
Peak theoretical throughput by precision type
| Precision | Peak TFLOPS | Efficiency |
|---|---|---|
| INT8 | 624.0 | 2.496 TFLOPS/W |
| INT4 | 1248.0 | 4.992 TFLOPS/W |
| FP16 | 312.0 | 1.248 TFLOPS/W |
| BF16 | 312.0 | 1.248 TFLOPS/W |
| TF32 | 156.0 | 0.624 TFLOPS/W |
| FP32 | 19.5 | 0.078 TFLOPS/W |
| FP64 | 9.7 | 0.039 TFLOPS/W |
FP16 Efficiency
1.248 TFLOPS/W
312.0 TFLOPS / 250W
FP32 Efficiency
0.078 TFLOPS/W
19.5 TFLOPS / 250W
Power Specifications
TDP
250 W
Max Power
288 W
Power Connector
PCIe 8-pin
Cooling
Air
Memory Specifications
Capacity
40 GB
Type
--
Bandwidth
1555 GB/s
Interface
--
Hardware & Design
Form Factor
PCIe
Architecture
Ampere
Process Node
7nm
Launch Year
2020
Variant
PCIe 40GB
Market Segment
Professional
Documentation & Resources
Common Use Cases
LLM Training LLM Inference Scientific Computing HPC Workloads
The NVIDIA A100 is optimized for high-performance computing tasks with Ampere architecture delivering 19.5 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
Systems Using This GPU
Pre-configured systems featuring the NVIDIA A100
| System | Peak Performance |
|---|---|
NVIDIA DGX A100 320GB DGX · rack · 2020 | 4.99 PFLOPS |
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
Loading verification...
No spam, unsubscribe anytime.