Specifications Compared
| Spec | A40 | L40 |
|---|---|---|
| TDP | 300W | 300W |
| VRAM | 48 GB | 48 GB |
| CUDA Cores | 10,752 | 18,176 |
| FP8 (dense) | Not published | 362 TFLOPS |
| Memory Type | GDDR6 | GDDR6 |
| Architecture | Ampere | Ada Lovelace |
| FP16 (dense) | 149.7 TFLOPS | 181.05 TFLOPS |
| Form Factors | PCIe | PCIe |
| INT8 (dense) | 299.3 TOPS | 362 TOPS |
| Interconnect | NVLink, PCIe 4.0 | PCIe 4.0 |
| Tensor Cores | 336 | 568 |
| FP32 Performance | 37.4 TFLOPS | 90.5 TFLOPS |
| Memory Bandwidth | 696 GB/s | 864 GB/s |
| FP8 (with sparsity) | Not published | 724 TFLOPS |
| FP16 (with sparsity) | 299.4 TFLOPS | 362.1 TFLOPS |
| INT8 (with sparsity) | 598.6 TOPS | 724 TOPS |
Performance Analysis
The FP16 dense performance ratio favors the L40 by the factor of 181.05 over 149.7. This difference affects training throughput where dense FP16 operations dominate. Memory bandwidth of 864 GB/s on the L40 versus 696 GB/s on the A40 permits larger batch sizes during inference tasks that rely on dense FP16 calculations. For sparse operations the L40 reaches 362.1 TFLOPS in FP16 with sparsity against 299.4 TFLOPS on the A40. The FP32 dense figure of 90.5 TFLOPS on the L40 exceeds the A40 value of 37.4 TFLOPS and influences scientific computing kernels that require full precision. These spec deltas translate to measurable differences in iteration times for models that stay within the shared 48 GB memory capacity.
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.
A40
1 provider in stock, 1 offer. All A40 offers, price history and alerts
L40
| Provider | Region | GPUs | Per GPU / hr | Instance / hr | Deploy |
|---|---|---|---|---|---|
| ThunderCompute | USA | 1 | $0.79 | — | Deploy |
| Massed Compute | us-central-3 | 4 | $0.86 | $3.44 | Deploy |
| QuantaCloud | us-midwest-2 | 4 | $0.94 | $3.75 | Deploy |
| Hyperstack | CANADA-1 | 2 | $1.00 | $2.00 | Deploy |
4 providers in stock, 17 offers (cheapest per provider shown). All L40 offers, price history and alerts
Notify me when A40 drops below a price
One email when the cheapest in-stock on-demand price per GPU-hour falls below your threshold. Cheapest right now: $0.52/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 A40
The A40 serves workloads that require NVLink interconnect for multi GPU scaling within the 48 GB memory envelope. Its 299.4 TFLOPS FP16 with sparsity rating supports specific sparse inference patterns where the 696 GB/s bandwidth remains sufficient. The 2020 Ampere design fits deployments that prioritize established PCIe configurations over newer FP8 capabilities.
When to Choose the L40
Its 90.5 TFLOPS FP32 rating accelerates precision sensitive kernels compared to the A40 37.4 TFLOPS value. The 2023 architecture supplies 362 TOPS dense INT8 for inference pipelines that operate without sparsity.
Use Cases
The L40 supplies 181.05 TFLOPS dense FP16 against 149.7 TFLOPS on the A40 for faster iteration within 48 GB memory.
The L40 864 GB/s bandwidth supports larger batches than the A40 696 GB/s during dense FP16 inference.
The L40 362.1 TFLOPS FP16 with sparsity exceeds the A40 299.4 TFLOPS for sparse fine tuning workloads.
The L40 362 TOPS dense INT8 rating accelerates diffusion steps compared to the A40 299.3 TOPS.
The L40 90.5 TFLOPS FP32 dense performance surpasses the A40 37.4 TFLOPS for full precision calculations.
Frequently Asked Questions
How does FP16 performance compare between A40 and L40?▾
The A40 provides 149.7 TFLOPS dense FP16 while the L40 provides 181.05 TFLOPS dense FP16. The A40 reaches 299.4 TFLOPS FP16 with sparsity and the L40 reaches 362.1 TFLOPS FP16 with sparsity.
What memory bandwidth do the A40 and L40 offer?▾
The A40 lists 696 GB/s memory bandwidth. The L40 lists 864 GB/s memory bandwidth. Both GPUs maintain 48 GB GDDR6 capacity.
Do the A40 and L40 support the same interconnect?▾
The A40 includes NVLink interconnect along with PCIe 4.0. The L40 supports only PCIe 4.0 interconnect.
What FP32 performance do these GPUs deliver?▾
The A40 delivers 37.4 TFLOPS FP32. The L40 delivers 90.5 TFLOPS FP32. Both operate at 300 W TDP.
How do the INT8 ratings differ?▾
The A40 lists 299.3 TOPS dense INT8 and 598.6 TOPS INT8 with sparsity. The L40 lists 362 TOPS dense INT8 and 724 TOPS INT8 with sparsity.
Which is cheaper to rent, the A40 or the L40?▾
Cloud rental prices for both the A40 and L40 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 A40 have compared to the L40?▾
The A40 has 48 GB of GDDR6 memory. The L40 has 48 GB of GDDR6 memory.
Can I find A40 and L40 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 A40 and the L40?▾
The A40 uses the Ampere architecture (2020) while the L40 uses Ada Lovelace (2023). The L40 delivers 1.2x the dense FP16 throughput (181.05 vs 149.7 TFLOPS, both without sparsity) and 1.2x 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 A40 and the L40. 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 A40Every current offer by provider, daily price history and a price alert.
- Rent L40Every 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.