NVIDIA A40 48GB
Performance
Peak theoretical throughput by precision type
| Precision | Dense | 2:4 Sparse |
|---|---|---|
FP64 | No verified data available | Structured sparsity is a tensor-core feature; this vector precision has no sparse form |
FP32 32-bit floating point | 37TFLOPS | Structured sparsity is a tensor-core feature; this vector precision has no sparse form |
TF32 TensorFloat-32 | 75TFLOPS | 150TFLOPS |
BF16 Brain Float 16 | 150TFLOPS | 299TFLOPS |
FP16 16-bit floating point | 150TFLOPS | 299TFLOPS |
FP8 | No verified data available | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
FP6 | No verified data available | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
FP4 | No verified data available | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
INT8 8-bit integer | 299TOPS | 599TOPS |
INT4 4-bit integer | 599TOPS | 1,197TOPS |
Dense peak divided by accelerator TDP. Board power only: excludes host CPUs, networking, cooling and facility overhead. TDP for this part: 300 W.
The A40 48GB in the GPU landscape
Peak FP16 TFLOPS (dense) against TDP, single-GPU parts tracked by Flopper
Higher and further left is better: more half-precision throughput for less power.
Specifications
Architecture
Ampere
Form Factor
PCIe
Launch Year
2021
Process Node
No verified data available
Memory
48 GB GDDR6
Bandwidth
696 GB/s
TDP
300 W
Max power (Flopper estimate)
~345 W est. Flopper estimate: 300 W TDP x 1.15. The vendor publishes no maximum board power for this part.
Spec Confidence
Official
Full Specifications
| Memory | |
|---|---|
| Memory | 48 GB |
| Memory Type | GDDR6 |
| Bandwidth | 696 GB/s |
| Interface Width | 384-bit |
| Power & Thermal | |
| TDP | 300 W |
| Max power (Flopper estimate) | ~345 W est. |
| Enterprise Features | |
| Sparsity | Yes |
| General | |
| Form Factor | PCIe |
| Architecture | Ampere |
| Process Node | No verified data available |
| Launch Year | 2021 |
Datasheet & Resources
Data Provenance
Every figure traced to a source
Primary Source
- Document
- —
- Publisher
- NVIDIA
- Published
- No verified data available
Data Quality
- Spec confidence
- Official
- Clock basis
- Boost
- Core precisions with figures
- 5 of 9
- Normalization
- All values in TFLOPS
Rent this GPU
$0.35 / GPU-hr
Lowest of 4 live listings across 3 providers· median $0.59
Median of each provider's cheapest on-demand listing, so a provider with many configurations counts once.
Price history and every provider on the pricing page. Some links are affiliate links.
Similar GPUs
Frequently Asked Questions
How many TFLOPS does the NVIDIA A40 have?
The NVIDIA A40 delivers 37 TFLOPS FP32 and 150 TFLOPS FP16 at peak. Flopper does not currently have a verified FP8 throughput figure for it.
What is the power consumption of the NVIDIA A40?
The NVIDIA A40 has a TDP (Thermal Design Power) rating of 300 watts.
How much memory does the NVIDIA A40 have?
The NVIDIA A40 is equipped with 48 GB of memory with 696 GB/s of memory bandwidth.
What architecture is the NVIDIA A40 based on?
The NVIDIA A40 is based on the Ampere architecture, launched in 2021.
How much does the NVIDIA A40 cost to rent?
The NVIDIA A40 rents from $0.35 per GPU-hour across 3 cloud providers, the lowest of 4 live listings Flopper tracks. The pricing page lists every provider and the price history.
When did the NVIDIA A40 come out?
The NVIDIA A40 launched in 2021 on the Ampere architecture.
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
No spam, unsubscribe anytime.