Specifications Compared
| Spec | A100 | A40 |
|---|---|---|
| TDP | 400W | 300W |
| VRAM | 40-80 GB | 48 GB |
| CUDA Cores | 6,912 | 10,752 |
| Memory Type | HBM2e | GDDR6 |
| Architecture | Ampere | Ampere |
| FP16 (dense) | 312 TFLOPS | 149.7 TFLOPS |
| Form Factors | SXM4, PCIe | PCIe |
| INT8 (dense) | 624 TOPS | 299.3 TOPS |
| Interconnect | NVLink, PCIe 4.0, InfiniBand | NVLink, PCIe 4.0 |
| Tensor Cores | 432 | 336 |
| FP32 Performance | 19.5 TFLOPS | 37.4 TFLOPS |
| FP64 Performance | 9.7 TFLOPS | Not published |
| Memory Bandwidth | 2,039 GB/s | 696 GB/s |
| FP16 (with sparsity) | 624 TFLOPS | 299.4 TFLOPS |
| INT8 (with sparsity) | 1,248 TOPS | 598.6 TOPS |
Performance Analysis
Dense FP16 throughput on the A100 reaches 312 TFLOPS while the A40 reaches 149.7 TFLOPS so the A100 supplies more than twice the dense FP16 operations per second. The same ratio holds for dense FP16 with sparsity where the A100 lists 624 TFLOPS against 299.4 TFLOPS on the A40. Higher memory bandwidth of 2039 GB/s on the A100 versus 696 GB/s on the A40 permits larger batch sizes during training before memory capacity becomes the limit. FP32 figures favor the A40 at 37.4 TFLOPS compared with 19.5 TFLOPS on the A100 which shifts suitability toward workloads that remain in single precision. INT8 dense performance follows the same pattern with 624 TOPS on the A100 versus 299.3 TOPS on the A40.
Current On-Demand Offers
Cheapest secure on-demand offer per provider that is reported in stock right now, per GPU per hour; multi-GPU instances show the whole-instance price alongside. Deploy opens the DeployGPU console for providers on the platform, otherwise the provider's own site. Stock is rechecked every minute.
A100
| Provider | Region | GPUs | Per GPU / hr | Instance / hr | Deploy |
|---|---|---|---|---|---|
| Vast.ai | Slovenia, SI | 1 | $0.67 | — | Deploy |
| LeaderGPU | The Netherlands | 8 | $0.68 | $5.42 | Deploy |
| ThunderCompute | USA | 2 | $1.09 | $2.18 | Deploy |
| Hyperstack | CANADA-1 | 1 | $1.35 | — | Deploy |
| Massed Compute | us-central-3 | 1 | $1.38 | — | Deploy |
11 providers in stock, 42 offers (cheapest per provider shown). All A100 offers, price history and alerts
A40
1 provider in stock, 1 offer. All A40 offers, price history and alerts
Notify me when A100 drops below a price
One email when the cheapest in-stock on-demand price per GPU-hour falls below your threshold. Cheapest right now: $0.67/GPU-hr.
QuantaCloud
Comparing A100 providers? We broker across all of them.
Need 16+ A100s reserved for fine-tuning, simulation, or production inference? We quote volume pricing across multiple data center partners: one quote at partner rates, 24h turnaround.
When to Choose the A100
The A100 suits LLM training workloads that require the 312 TFLOPS dense FP16 rating and 2039 GB/s memory bandwidth to sustain large batch sizes. Its 624 TOPS dense INT8 rating further supports high throughput inference on quantized models where bandwidth prevents stalls.
When to Choose the A40
The A40 fits scientific computing tasks that rely on its 37.4 TFLOPS FP32 rating and 300 W TDP. Its 48 GB GDDR6 capacity also supports inference pipelines that stay within single precision limits without needing the higher bandwidth of the A100.
Use Cases
The A100 supplies 312 TFLOPS dense FP16 and 2039 GB/s bandwidth to maintain large batches during training.
The A100 lists 624 TOPS dense INT8 which exceeds the 299.3 TOPS rating of the A40 for quantized inference.
Dense FP16 performance of 312 TFLOPS on the A100 enables faster gradient updates than the 149.7 TFLOPS on the A40.
The A40 provides 37.4 TFLOPS FP32 which exceeds the 19.5 TFLOPS rating of the A100 for diffusion workloads that remain in single precision.
FP32 throughput of 37.4 TFLOPS on the A40 combined with its 300 W TDP suits precision sensitive simulations.
Frequently Asked Questions
What is the memory bandwidth difference between A100 and A40?▾
The A100 reaches 2039 GB/s while the A40 reaches 696 GB/s. This gap directly affects maximum batch size before memory stalls occur during dense FP16 operations.
How does FP16 performance compare between the two GPUs?▾
Dense FP16 on the A100 measures 312 TFLOPS against 149.7 TFLOPS on the A40. The same ratio applies to FP16 with sparsity at 624 TFLOPS versus 299.4 TFLOPS.
Which GPU offers higher FP32 throughput?▾
The A40 lists 37.4 TFLOPS FP32 compared with 19.5 TFLOPS on the A100. Workloads that stay in single precision therefore favor the A40.
What are the TDP ratings for A100 and A40?▾
The A100 carries a 400 W TDP while the A40 carries a 300 W TDP. Power limited deployments must account for this 100 W difference.
Does the A100 support NVLink?▾
Both GPUs list NVLink interconnect support. The A100 additionally offers SXM4 and PCIe form factors while the A40 is limited to PCIe.
What INT8 performance numbers are published?▾
Dense INT8 reaches 624 TOPS on the A100 versus 299.3 TOPS on the A40. With sparsity the figures become 1,248 TOPS and 598.6 TOPS respectively.
Which is cheaper to rent, the A100 or the A40?▾
Cloud rental prices for both the A100 and A40 vary by provider, configuration, and availability. This page shows live pricing from 25+ providers updated every 60 seconds. Scroll to the Live Cloud Pricing section to compare current rates.
How much VRAM does the A100 have compared to the A40?▾
The A100 has 40 to 80 GB of HBM2e memory. The A40 has 48 GB of GDDR6 memory.
Can I find A100 and A40 GPUs available to rent right now?▾
Yes. This page shows real-time availability across 25+ cloud GPU providers. The Live Cloud Pricing section displays only in-stock offers with current pricing.
What is the main difference between the A100 and the A40?▾
The A100 uses the Ampere architecture (2020) while the A40 uses Ampere (2020). The A100 delivers 2.1x the dense FP16 throughput (312 vs 149.7 TFLOPS, both without sparsity) and 2.9x the memory bandwidth of the A40.
Rent these GPUs
Each GPU page lists every current on-demand offer by provider, per GPU-hour, updated every minute.
Related comparisons
How this page is made
- Specifications come from the NVIDIA, AMD and Intel datasheets for the A100 and the A40. Dense and sparse throughput are listed separately.
- Prices and availability are live: every offer shown is an in-stock, on-demand listing from a provider we track, rechecked every minute.
- The written comparison (overview, performance notes, when to choose each card, verdict, use cases and FAQ) was drafted with an AI model from the specification table above and passed an automated check that rejects any figure not in that table. It was last generated on . It contains no prices; those are always read live.
- Read how we collect the data, or report an error on this page.
Next steps
- Rent A100Every current offer by provider, daily price history and a price alert.
- Rent A40Every current offer by provider, daily price history and a price alert.
- GPU price indexHow on-demand prices for the major GPUs have moved, updated daily.
- All GPUsEvery GPU we track, with current lows and provider counts.