Specifications Compared
| Spec | V100 | A100 |
|---|---|---|
| TDP | 300W | 400W |
| VRAM | 16-32 GB | 40-80 GB |
| CUDA Cores | 5,120 | 6,912 |
| Memory Type | HBM2 | HBM2e |
| Architecture | Volta | Ampere |
| FP16 (dense) | 125 TFLOPS | 312 TFLOPS |
| Form Factors | SXM2, PCIe | SXM4, PCIe |
| INT8 (dense) | Not published | 624 TOPS |
| Interconnect | NVLink, PCIe 3.0 | NVLink, PCIe 4.0, InfiniBand |
| Tensor Cores | 640 | 432 |
| FP32 Performance | 15.7 TFLOPS | 19.5 TFLOPS |
| FP64 Performance | 7.8 TFLOPS | 9.7 TFLOPS |
| Memory Bandwidth | 900 GB/s | 2,039 GB/s |
| FP16 (with sparsity) | Not published | 624 TFLOPS |
| INT8 (with sparsity) | Not published | 1,248 TOPS |
Performance Analysis
The FP16 dense performance ratio shows the A100 at 312 TFLOPS versus the V100 at 125 TFLOPS. This gap means training runs that rely on dense FP16 operations complete in fewer steps on the A100. The FP32 figures stand at 19.5 TFLOPS for the A100 and 15.7 TFLOPS for the V100 so mixed precision pipelines gain most of their acceleration from the FP16 dense path. Memory bandwidth of 2039 GB/s on the A100 compared with 900 GB/s on the V100 permits larger batch sizes before the workload becomes memory bound. When sparsity is available the A100 reaches 624 TFLOPS FP16 with sparsity while the V100 provides no sparsity figure for comparison.
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.
V100
| Provider | Region | GPUs | Per GPU / hr | Instance / hr | Deploy |
|---|---|---|---|---|---|
| Ori | lille-1 | 2 | $0.83 | $1.66 | Deploy |
| Paperspace | ny2 | 1 | $2.30 | — | Deploy |
2 providers in stock, 72 offers (cheapest per provider shown). All V100 offers, price history and alerts
A100
| Provider | Region | GPUs | Per GPU / hr | Instance / hr | 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-2 | 1 | $1.38 | — | Deploy |
| QuantaCloud | us-midwest-2 | 8 | $1.49 | $11.92 | Deploy |
10 providers in stock, 28 offers (cheapest per provider shown). All A100 offers, price history and alerts
Notify me when V100 drops below a price
One email when the cheapest in-stock on-demand price per GPU-hour falls below your threshold. Cheapest right now: $0.83/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 V100
The V100 suits environments that must stay within 300W TDP limits. Its 16 to 32 GB HBM2 configuration remains adequate for inference tasks that fit inside that capacity and that do not require the 312 TFLOPS dense FP16 rate of newer hardware.
When to Choose the A100
The A100 suits workloads that need 40 to 80 GB HBM2e capacity together with 2039 GB/s bandwidth. Its 312 TFLOPS dense FP16 and 624 TFLOPS FP16 with sparsity accelerate both training and inference relative to the V100 125 TFLOPS dense FP16 baseline.
Use Cases
The A100 supplies 312 TFLOPS dense FP16 against the V100 125 TFLOPS dense FP16 and doubles memory bandwidth to 2039 GB/s.
The A100 624 TFLOPS FP16 with sparsity and 40 to 80 GB capacity allow larger batch inference than the V100 limits.
The A100 19.5 TFLOPS FP32 combined with higher memory bandwidth supports fine-tuning iterations more efficiently than the V100 15.7 TFLOPS FP32.
The A100 312 TFLOPS dense FP16 and 2039 GB/s bandwidth accommodate diffusion model memory demands beyond V100 125 TFLOPS dense FP16 and 900 GB/s.
The V100 15.7 TFLOPS FP32 at 300W TDP matches some workloads while the A100 19.5 TFLOPS FP32 offers headroom when power budget allows 400W.
Frequently Asked Questions
What is the FP16 dense performance difference between V100 and A100?▾
The V100 lists 125 TFLOPS dense FP16 while the A100 lists 312 TFLOPS dense FP16. This difference directly affects training step time for models that use dense FP16 arithmetic.
Which GPU has higher FP32 throughput?▾
The A100 reaches 19.5 TFLOPS FP32 compared with the V100 15.7 TFLOPS FP32. The margin favors the A100 for workloads that remain in FP32 precision.
What TDP values are listed for these GPUs?▾
The V100 TDP is 300W and the A100 TDP is 400W. Systems with strict power envelopes may therefore retain the V100.
Does the A100 support sparsity in FP16?▾
The A100 lists 624 TFLOPS FP16 with sparsity. The V100 publishes no sparsity figure so only dense comparisons apply between the two GPUs.
Which is cheaper to rent, the V100 or the A100?▾
Cloud rental prices for both the V100 and A100 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 V100 have compared to the A100?▾
The V100 has 16 to 32 GB of HBM2 memory. The A100 has 40 to 80 GB of HBM2e memory.
Can I find V100 and A100 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 V100 and the A100?▾
The V100 uses the Volta architecture (2017) while the A100 uses Ampere (2020). The A100 delivers 2.5x the dense FP16 throughput (312 vs 125 TFLOPS, both without sparsity) and 2.3x the memory bandwidth of the V100.
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 V100 and the A100. 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 V100Every current offer by provider, daily price history and a price alert.
- Rent A100Every 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.