Nvidia A40
Data center GPU for large-batch inference and professional visualization.
Compare vs other GPUs →Key Specifications
Compare Cloud Provider Prices
For smaller projects, the A40 is available around $0.22/hr per GPU (12-month reservation). This tier offers a 89% discount compared to the higher end of the market ($2.14/hr).
How we sort results
Default ranking
Our algorithm weighs five factors to find the most relevant matches for you:
- Price: We blend total hourly price with price-per-GPU to balance affordability and value.
- Specs: We favor offers with higher CPU, RAM, and GPU Memory.
- Billing: We favor on-demand billing for simplicity and flexibility over spot instances, reservations, and custom quotes.
- Location: We blend both datacenter proximity and provider HQ location. Datacenter location matters for latency, while HQ location matters for compliance and support.
- Provider diversity: Each time a provider appears in the list, their subsequent offerings are ranked a little lower, so one provider's offerings don't crowd out the top positions.
Sorting and filtering
Click any column header to sort by that column. Use the filters above the table to narrow results by billing type, GPU count, vCPUs, or RAM. Custom sorting resets the default relevance ranking.
Transparency and funding
Ads and sponsors: Any paid placements are fixed at the top and clearly labeled as sponsored content.
Affiliates: Any affiliate links will be indicated to you as well. We may earn a commission if you click them, but this never influences the ranking order.
| Provider | GPUs | Total VRAM | vCPUs | RAM | Billing | $/GPU/h | Total/h | Availability | |
|---|---|---|---|---|---|---|---|---|---|
Runpod
Our sponsor
|
1x A40 1x A40 48GB (secure-cloud) | 48GB | 9 | 48GB | On-Demand Pay-as-you-go pricing. No term commitments. | $0.44 | $0.44 | Available | View |
Crusoe
|
1x A40 1x A40 48GB (a40.1x) | 48GB | 6 | 60GB | On-Demand Pay-as-you-go pricing. No term commitments. | $0.90 | $0.90 | Unknown | View |
|
Runpod
|
2x A40 2x A40 48GB (secure-cloud) | 96GB | 18 | 96GB | On-Demand Pay-as-you-go pricing. No term commitments. | $0.44 | $0.88 | Available | View |
Vultr
|
1x A40 1x A40 48GB | 48GB | 24 | 120GB | On-Demand Pay-as-you-go pricing. No term commitments. | $1.71 | $1.71 | Unknown | View |
|
Runpod
|
4x A40 4x A40 48GB (secure-cloud) | 192GB | 36 | 192GB | On-Demand Pay-as-you-go pricing. No term commitments. | $0.44 | $1.76 | Available | View |
Database Mart
|
1x A40 1x A40 48GB (Enterprise Dedicated GPU Server - A40) | 48GB | 36 | 256GB | Reserved (1mo) Reserved capacity with a 1-month minimum commitment. | $0.76 | $0.76 | Unknown | View |
Vast.ai
|
2x A40 2x A40 46GB (30818884) | 92GB | 256 | 967GB | On-Demand Pay-as-you-go pricing. No term commitments. | $0.29 | $0.57 | Available | View |
|
Runpod
|
8x A40 8x A40 48GB (secure-cloud) | 384GB | 72 | 384GB | On-Demand Pay-as-you-go pricing. No term commitments. | $0.44 | $3.52 | Available | View |
|
Vultr
|
4x A40 4x A40 48GB | 192GB | 96 | 480GB | On-Demand Pay-as-you-go pricing. No term commitments. | $1.71 | $6.85 | Unknown | View |
Exoscale
|
1x A40 1x A40 48GB | 48GB | 12 | 56GB | On-Demand Pay-as-you-go pricing. No term commitments. | $2.14 | $2.14 | Unknown | View |
|
Crusoe
|
1x A40 1x A40 48GB (a40.1x (Spot)) | 48GB | 6 | 60GB | Spot Variable spot pricing with potential interruptions. | $0.40 | $0.40 | Unknown | View |
|
Exoscale
|
2x A40 2x A40 48GB | 96GB | 24 | 120GB | On-Demand Pay-as-you-go pricing. No term commitments. | $2.13 | $4.27 | Unknown | View |
|
Runpod
|
1x A40 1x A40 48GB (secure-cloud) | 48GB | 9 | 48GB | Reserved (12mo) Reserved capacity with a 1-year minimum commitment. | $0.22 | $0.22 | Available | View |
|
Runpod
|
1x A40 1x A40 48GB (secure-cloud) | 48GB | 9 | 48GB | Reserved (6mo) Reserved capacity with a 6-month minimum commitment. | $0.31 | $0.31 | Available | View |
|
Vast.ai
|
2x A40 2x A40 46GB (30818884) | 92GB | 256 | 967GB | Spot Variable spot pricing with potential interruptions. | $0.23 | $0.47 | Available | View |
|
Exoscale
|
4x A40 4x A40 48GB | 192GB | 48 | 224GB | On-Demand Pay-as-you-go pricing. No term commitments. | $2.13 | $8.54 | Unknown | View |
|
Runpod
|
1x A40 1x A40 48GB (secure-cloud) | 48GB | 9 | 48GB | Reserved (3mo) Reserved capacity with a 3-month minimum commitment. | $0.33 | $0.33 | Available | View |
|
Runpod
|
2x A40 2x A40 48GB (secure-cloud) | 96GB | 18 | 96GB | Reserved (12mo) Reserved capacity with a 1-year minimum commitment. | $0.22 | $0.44 | Available | View |
|
Runpod
|
2x A40 2x A40 48GB (secure-cloud) | 96GB | 18 | 96GB | Reserved (6mo) Reserved capacity with a 6-month minimum commitment. | $0.31 | $0.62 | Available | View |
|
Exoscale
|
8x A40 8x A40 48GB | 384GB | 96 | 448GB | On-Demand Pay-as-you-go pricing. No term commitments. | $2.13 | $17.06 | Unknown | View |
| No offerings matching your filters. | |||||||||
Heads up: We do our best to keep these prices accurate. However, cloud costs may fluctuate based on region, usage, and other factors not listed here. These are estimates based on common setups and are for informational purposes only. Always verify current rates with the provider before provisioning.
Frequently Asked Questions
Why choose the A40?
48GB GDDR6 with Ampere Tensor Cores. Large VRAM for inference on bigger models. Supports both AI and professional visualization workloads.
When is the A40 not a good fit?
No FP8 support. PCIe-only. For pure AI inference, the L40S offers better performance with Ada Lovelace architecture at similar or lower cost.
Are A40 prices going up or down?
On-demand pricing has increased by about 14% since June 2025, from $1.61 to $1.84/hr per GPU.
What size AI models can the A40 run?
With 48GB of VRAM, the A40 can typically run models up to about 30B parameters in FP16, or 70B-class models in 4-bit quantized form for inference.
How much VRAM does the A40 have?
The A40 has 48GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.
What is the A40's memory bandwidth?
The A40 has 696 GB/s of memory bandwidth. Higher bandwidth helps with faster data transfer between GPU memory and compute cores.
What data types does the A40 support?
The A40 supports 6 precision formats. Training: BF16, FP16, TF32, FP32. Inference: INT4, INT8.
Does the A40 support NVLink?
No. The A40 is a PCIe-only GPU with no NVLink, so it is better suited to single-GPU inference and smaller-scale workloads than large distributed training jobs.
How much does the A40 cost per hour?
A40 pricing currently ranges from $0.22/hr to $2.14/hr per GPU, depending on the provider, instance type, and billing model.
How much does the A40 cost per month?
At 720 hours per month, one A40 can cost between $158.40 to $1,540.80 per month, depending on the provider. Reserved and spot pricing can lower that further.
Which cloud providers offer the A40?
The A40 is available from 7 cloud providers, including Runpod, Exoscale, Crusoe. Pricing and availability vary by region and billing model.
Can I rent the A40 in the cloud?
Yes. We currently track 28 A40 listings across 7 cloud providers:
Billing type Listings Avg $/GPU/hr On-demand 13 $1.24/hr Reserved 13 $0.32/hr Spot 2 $0.32/hr
Technical Specifications
| GPU architecture | NVIDIA Ampere architecture |
| GPU memory | 48 GB GDDR6 |
| 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.
Alternatives to Nvidia A40
Last updated