Specifications Compared
| Spec | A10 | L40S |
|---|---|---|
| TDP | 150W | 350W |
| VRAM | 24 GB | 48 GB |
| CUDA Cores | 9,216 | 18,176 |
| FP8 (dense) | Not published | 733 TFLOPS |
| Memory Type | GDDR6 | GDDR6 |
| Architecture | Ampere | Ada Lovelace |
| FP16 (dense) | 125 TFLOPS | 362.05 TFLOPS |
| Form Factors | PCIe | PCIe |
| INT8 (dense) | 250 TOPS | 733 TOPS |
| Interconnect | PCIe 4.0 | PCIe 4.0 |
| Tensor Cores | 288 | 568 |
| FP32 Performance | 31.2 TFLOPS | 91.6 TFLOPS |
| FP64 Performance | Not published | 1.4 TFLOPS |
| Memory Bandwidth | 600 GB/s | 864 GB/s |
| FP8 (with sparsity) | Not published | 1,466 TFLOPS |
| FP16 (with sparsity) | 250 TFLOPS | 733 TFLOPS |
| INT8 (with sparsity) | 500 TOPS | 1,466 TOPS |
Performance Analysis
The FP16 dense performance reaches 125 TFLOPS on the A10 and 362.05 TFLOPS on the L40S. The FP16 figure with sparsity reaches 250 TFLOPS on the A10 and 733 TFLOPS on the L40S. This delta indicates the L40S processes larger batch sizes during training or inference steps that rely on dense FP16 operations. Memory bandwidth of 600 GB/s on the A10 versus 864 GB/s on the L40S limits sustained data movement on the A10 when models exceed 24 GB. The FP32 performance of 31.2 TFLOPS on the A10 compared with 91.6 TFLOPS on the L40S shows the L40S completes general matrix operations in fewer cycles for both dense and sparse paths.
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.
A10
| Provider | Region | GPUs | Per GPU / hr | Instance / hr | Deploy |
|---|---|---|---|---|---|
| LeaderGPU | The Netherlands | 10 | $0.37 | $3.71 | Deploy |
| Lambda Labs | us-west-1 | 1 | $1.29 | — | Deploy |
2 providers in stock, 3 offers (cheapest per provider shown). All A10 offers, price history and alerts
L40S
| Provider | Region | GPUs | Per GPU / hr | Instance / hr | Deploy |
|---|---|---|---|---|---|
| Massed Compute | us-central-2 | 4 | $0.97 | $3.88 | Deploy |
| QuantaCloud | us-midwest-2 | 2 | $1.09 | $2.18 | Deploy |
| Ori | lille-4 | 1 | $1.55 | — | Deploy |
| Lyceum | Europe | 1 | $1.99 | — | Deploy |
4 providers in stock, 13 offers (cheapest per provider shown). All L40S offers, price history and alerts
Notify me when A10 drops below a price
One email when the cheapest in-stock on-demand price per GPU-hour falls below your threshold. Cheapest right now: $0.37/GPU-hr.
QuantaCloud
Comparing providers? We broker across all of them.
Stop tab-switching between pricing pages. Tell us what you need, 16+ GPUs reserved or cluster capacity, and we return one quote at partner rates within 24 hours.
When to Choose the A10
The A10 GPU suits deployments where TDP must remain at 150W and VRAM demand stays within 24 GB. Workloads that operate on INT8 dense at 250 TOPS benefit from the lower power envelope without requiring the 350W TDP of the L40S.
When to Choose the L40S
The L40S GPU suits deployments where FP16 dense performance of 362.05 TFLOPS and 48 GB memory allow larger models or higher throughput. Scientific workloads that scale with 91.6 TFLOPS FP32 and 864 GB/s bandwidth select the L40S over the A10.
Use Cases
The L40S provides 48 GB GDDR6 memory and 362.05 TFLOPS dense FP16 compared with 24 GB and 125 TFLOPS dense FP16 on the A10.
The L40S supplies 733 TFLOPS FP16 with sparsity and 864 GB/s bandwidth while the A10 supplies 250 TFLOPS FP16 with sparsity and 600 GB/s bandwidth.
The L40S delivers 733 TOPS INT8 dense and 48 GB memory against 250 TOPS INT8 dense and 24 GB memory on the A10.
The L40S offers 362.05 TFLOPS dense FP16 and 91.6 TFLOPS FP32 while the A10 offers 125 TFLOPS dense FP16 and 31.2 TFLOPS FP32.
The L40S provides 91.6 TFLOPS FP32 and 864 GB/s bandwidth compared with 31.2 TFLOPS FP32 and 600 GB/s bandwidth on the A10.
Frequently Asked Questions
What is the FP32 performance difference between the A10 and L40S?▾
The A10 reaches 31.2 TFLOPS FP32. The L40S reaches 91.6 TFLOPS FP32.
Which GPU has higher memory bandwidth, the A10 or the L40S?▾
The A10 has 600 GB/s memory bandwidth. The L40S has 864 GB/s memory bandwidth.
What are the dense FP16 figures for the A10 and the L40S?▾
The A10 reaches 125 TFLOPS dense FP16. The L40S reaches 362.05 TFLOPS dense FP16.
How do the TDP values compare for the A10 and L40S?▾
The A10 has a TDP of 150W. The L40S has a TDP of 350W.
What INT8 dense performance does each GPU provide?▾
The A10 provides 250 TOPS INT8 dense. The L40S provides 733 TOPS INT8 dense.
Which is cheaper to rent, the A10 or the L40S?▾
Cloud rental prices for both the A10 and L40S 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 A10 have compared to the L40S?▾
The A10 has 24 GB of GDDR6 memory. The L40S has 48 GB of GDDR6 memory.
Can I find A10 and L40S 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 A10 and the L40S?▾
The A10 uses the Ampere architecture (2021) while the L40S uses Ada Lovelace (2023). The L40S delivers 2.9x the dense FP16 throughput (362.05 vs 125 TFLOPS, both without sparsity) and 1.4x the memory bandwidth of the A10.
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 A10 and the L40S. 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 A10Every current offer by provider, daily price history and a price alert.
- Rent L40SEvery 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.