Specifications Compared
| Spec | L40 | RTX 3090 |
|---|---|---|
| TDP | 300W | 350W |
| VRAM | 48 GB | 24 GB |
| CUDA Cores | 18,176 | 10,496 |
| FP8 (dense) | 362 TFLOPS | Not published |
| Memory Type | GDDR6 | GDDR6X |
| Architecture | Ada Lovelace | Ampere |
| FP16 (dense) | 181.05 TFLOPS | 71 TFLOPS |
| Form Factors | PCIe | PCIe |
| INT8 (dense) | 362 TOPS | 284 TOPS |
| Interconnect | PCIe 4.0 | NVLink, PCIe 4.0 |
| Tensor Cores | 568 | 328 |
| FP32 Performance | 90.5 TFLOPS | 35.6 TFLOPS |
| Memory Bandwidth | 864 GB/s | 936 GB/s |
| FP8 (with sparsity) | 724 TFLOPS | Not published |
| FP16 (with sparsity) | 362.1 TFLOPS | 142 TFLOPS |
| INT8 (with sparsity) | 724 TOPS | 568 TOPS |
Performance Analysis
The FP16 dense performance difference means the L40 executes matrix operations at 181.05 TFLOPS while the RTX 3090 reaches 71 TFLOPS for training and inference tasks. The FP32 figure of 90.5 TFLOPS on the L40 versus 35.6 TFLOPS on the RTX 3090 further widens the gap for precision sensitive workloads. Memory bandwidth of 864 GB/s on the L40 versus 936 GB/s on the RTX 3090 influences maximum batch sizes in memory bound scenarios where the RTX 3090 supports modestly larger batches per step.
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.
L40
| Provider | Region | GPUs | Per GPU / hr | Instance / hr | Deploy |
|---|---|---|---|---|---|
| ThunderCompute | USA | 1 | $0.79 | — | Deploy |
| RunPod | global | 1 | $0.82 | — | 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 | 4 | $1.00 | $4.00 | Deploy |
5 providers in stock, 16 offers (cheapest per provider shown). All L40 offers, price history and alerts
Notify me when L40 drops below a price
One email when the cheapest in-stock on-demand price per GPU-hour falls below your threshold. Cheapest right now: $0.79/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 L40
The L40 suits scenarios that require 48 GB VRAM for large model states and 181.05 TFLOPS FP16 dense throughput at 300W TDP. Users select the L40 when FP8 dense performance at 362 TFLOPS or INT8 dense performance at 362 TOPS accelerates quantized inference pipelines.
When to Choose the RTX 3090
The 350W TDP model handles tasks where the 24 GB VRAM capacity remains sufficient and higher bandwidth reduces data movement stalls.
Use Cases
The L40 supplies 181.05 TFLOPS FP16 dense and 48 GB VRAM to support larger training batches than the 71 TFLOPS FP16 dense and 24 GB on the RTX 3090.
The L40 achieves 362 TFLOPS FP8 dense to accelerate inference throughput beyond the RTX 3090 capabilities.
The L40 provides 181.05 TFLOPS FP16 dense and 48 GB VRAM for fine tuning large models that exceed 24 GB capacity on the RTX 3090.
The L40 delivers 90.5 TFLOPS FP32 to handle scientific simulations more efficiently than the 35.6 TFLOPS FP32 on the RTX 3090.
Frequently Asked Questions
What is the FP16 dense performance of each GPU?▾
The L40 reaches 181.05 TFLOPS FP16 dense and the RTX 3090 reaches 71 TFLOPS FP16 dense.
Which GPU has higher memory bandwidth?▾
The RTX 3090 provides 936 GB/s bandwidth compared to 864 GB/s on the L40.
How do the TDP ratings differ between the L40 and RTX 3090?▾
The L40 operates at 300W TDP while the RTX 3090 operates at 350W TDP.
Does the RTX 3090 support FP8 operations?▾
The RTX 3090 does not publish FP8 figures while the L40 publishes 362 TFLOPS FP8 dense.
What interconnect options exist on each card?▾
The L40 uses PCIe 4.0 while the RTX 3090 supports both NVLink and PCIe 4.0.
Which is cheaper to rent, the L40 or the RTX 3090?▾
Cloud rental prices for both the L40 and RTX 3090 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 L40 have compared to the RTX 3090?▾
The L40 has 48 GB of GDDR6 memory. The RTX 3090 has 24 GB of GDDR6X memory.
Can I find L40 and RTX 3090 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 L40 and the RTX 3090?▾
The L40 uses the Ada Lovelace architecture (2023) while the RTX 3090 uses Ampere (2020). The L40 delivers 2.5x the dense FP16 throughput (181.05 vs 71 TFLOPS, both without sparsity) while the RTX 3090 has 1.1x the memory bandwidth of the L40.
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 L40 and the RTX 3090. 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 L40Every current offer by provider, daily price history and a price alert.
- Rent RTX 3090Every 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.