Nvidia A40
Compare prices for Nvidia A40 across cloud providers
March 14, 2025 (updated)
Launched in 2020, the Nvidia A40 handles high-performance AI, rendering, and visualization workloads for enterprise environments. Nvidia emphasizes its suitability for complex tasks requiring robust computational power.
| Provider | GPUs | VRAM | vCPUs | RAM | Price/h | |
|---|---|---|---|---|---|---|
RunPod
|
1x A40 | 48GB | 9 | 50GB | $0.39 | Source |
Hyperstack
|
1x A40 | 48GB | 28 | 58GB | $0.50 | Source |
CUDO Compute
|
1x A40 | 48GB | 6 | 24GB | $0.53 | Launch |
FluidStack
|
1x A40 | 48GB | 32 | 128GB | $0.60 | Source |
Crusoe
|
1x A40 | 48GB | 6 | 60GB | $1.10 | Launch |
Vultr
|
1x A40 | 48GB | 24 | 120GB | $1.86 | Launch |
Replicate
|
1x A40 | 48GB | 4 | 16GB | $2.07 | Source |
Exoscale
|
1x A40 | 48GB | 12 | 56GB | $2.14 | Source |
|
Replicate
|
1x A40 | 48GB | 10 | 72GB | $2.61 | Source |
|
Exoscale
|
2x A40 | 96GB | 24 | 120GB | $4.27 | Source |
|
Replicate
|
2x A40 | 96GB | 20 | 144GB | $5.22 | Source |
|
Vultr
|
4x A40 | 192GB | 96 | 480GB | $7.44 | Launch |
|
Exoscale
|
4x A40 | 192GB | 48 | 224GB | $8.54 | Source |
|
Replicate
|
4x A40 | 192GB | 40 | 288GB | $10.44 | Source |
|
Exoscale
|
8x A40 | 384GB | 96 | 448GB | $17.06 | Source |
|
Replicate
|
8x A40 | 384GB | 48 | 680GB | $20.88 | Source |
Note: Prices are subject to change and may vary by region and other factors not listed here. For some GPUs, I include links to Shadeform (the sponsor) so you can check if they're available right now. I don’t earn a commission when you click on these links, but their monthly sponsorship helps me keep the site running.
Nvidia A40 specs
| GPU architecture | NVIDIA Ampere architecture |
| GPU memory | 48 GB GDDR6 with ECC |
| Memory bandwidth | 696 GB/s |
| Interconnect interface | NVIDIA® NVLink® 112.5 GB/s (bidirectional), PCIe Gen4: 64GB/s |
| NVIDIA Ampere architecture-based CUDA Cores | 10,752 |
| NVIDIA second-generation RT Cores | 84 |
| NVIDIA third-generation Tensor Cores | 336 |
| Peak FP32 TFLOPS (non-Tensor) | 37.4 |
| Peak FP16 Tensor TFLOPS with FP16 Accumulate | 149.7, 299.4* |
| Peak TF32 Tensor TFLOPS | 74.8, 149.6* |
| RT Core performance TFLOPS | 73.1 |
| Peak BF16 Tensor TFLOPS with FP32 Accumulate | 149.7, 299.4* |
| Peak INT8 Tensor TOPS, Peak INT4 Tensor TOPS | 299.3, 598.6, 598.7, 1,197.4 |
| Form factor | 4.4” (H) x 10.5” (L) dual slot |
| Display ports | 3x DisplayPort 1.4**, Supports NVIDIA Mosaic and Quadro® Sync |
| Max power consumption | 300 W |
| Power connector | 8-pin CPU |
| Thermal solution | Passive |
| Virtual GPU (vGPU) software support | NVIDIA vPC/vApps, NVIDIA RTX Virtual Workstation, NVIDIA Virtual Compute Server |
| vGPU profiles supported | See the Virtual GPU Licensing Guide |
| NVENC, NVDEC | 1x, 2x (includes AV1 decode) |
| Secure and measured boot with hardware root of trust | Yes (optional) |
| NEBS ready | Level 3 |
| Compute APIs | CUDA, DirectCompute, OpenCL™, OpenACC® |
| Graphics APIs | DirectX 12.07, Shader Model 5.17, OpenGL 4.68, Vulkan 1.18 |
| MIG support | No |
Source: official Nvidia A40 datasheet.