NVIDIA L40 (48GB GDDR6)
- 48GB GDDR6
- Memory
- 864GB/s
- Memory bandwidth
- Q4 2022
- Released
Price history
What it costs to rent a NVIDIA L40
The NVIDIA L40 is an Ada Lovelace-architecture inference NVIDIA GPU with 48GB of GDDR6 memory and 864 GB/s of memory bandwidth, released Q4 2022. It is built for inference, fine-tuning and mixed production workloads.
Renting one currently costs between $0.40 and $1.25 per GPU-hour across the 9 cloud providers we track, with a median of $0.95. The cheapest live rate is $0.40/hr from Vast.ai (Spot), and hosting one pays about $0.48/hr on RunPod while it is running. All rates below are normalised to a single GPU per hour so providers with multi-GPU nodes compare like-for-like.
Where to rent the L40
How we rank
The median is one provider, one vote: each provider contributes the median of its own on-demand listings, then the headline is the median across providers — so no single marketplace or hyperscaler sets the price alone. Below two providers we show the range instead.
Default order is relevance, weighed the way a buyer would: price per GPU first, lightly blended with total node cost; on-demand preferred over spot and reserved when prices are close; established clouds edged above marketplaces; in-stock, recently-checked offers above stale ones; and no single provider is allowed to crowd the top. Click a column to sort by the raw numbers. No paid placements — ranking is never sold.
| Provider | Tier | Config | Total VRAM | Billing | $ / GPU / hr | Total / hr | Availability | Go to provider |
|---|---|---|---|---|---|---|---|---|
 Vast.ai | GPU cloud | 1× L40 | 48GB | Spot | $0.40 | $0.40 | In stock | Rent |
| Vast.ai | GPU cloud | 1× L40 | 48GB | On-demand | $0.46 | $0.46 | In stock | Rent |
| Vast.ai | GPU cloud | 2× L40 | 96GB | Spot | $0.40 | $0.80 | In stock | Rent |
| Vast.ai | GPU cloud | 2× L40 | 96GB | On-demand | $0.46 | $0.91 | In stock | Rent |
| Vast.ai | GPU cloud | 3× L40 | 144GB | Spot | $0.40 | $1.20 | In stock | Rent |
| Vast.ai | GPU cloud | 3× L40 | 144GB | On-demand | $0.45 | $1.36 | In stock | Rent |
 RunPod | GPU cloud | 1× L40 | 48GB | Community | $0.69 | $0.69 | In stock | Rent |
 Thunder Compute | Marketplace | 1× L40 | 48GB | On-demand | $0.79 | $0.79 | Unknown | Rent |
 Massed Compute | GPU cloud | 1× L40 | 48GB | On-demand | $0.86 | $0.86 | In stock | Rent |
 Oblivus | Marketplace | 1× L40 | 48GB | Reserved | $0.74 | $0.74 | In stock | Rent |
| RunPod | GPU cloud | 2× L40 | 96GB | Community | $0.69 | $1.38 | In stock | Rent |
 Hyperstack | GPU cloud | 1× L40 | 48GB | Spot | $0.80 | $0.80 | Unknown | Rent |
| Thunder Compute | Marketplace | 2× L40 | 96GB | On-demand | $0.79 | $1.58 | Unknown | Rent |
 Sesterce | GPU cloud | 1× L40 | 48GB | On-demand | $0.97 | $0.97 | In stock | Rent |
| Massed Compute | GPU cloud | 2× L40 | 96GB | On-demand | $0.86 | $1.72 | In stock | Rent |
| Oblivus | Marketplace | 2× L40 | 96GB | Reserved | $0.74 | $1.48 | In stock | Rent |
| Vast.ai | GPU cloud | 4× L40 | 192GB | Spot | $0.47 | $1.87 | In stock | Rent |
 Runcrate | Marketplace | from L40 | — | On-demand | $0.97 | — | Unknown | Rent |
| Massed Compute | GPU cloud | 4× L40 | 192GB | On-demand | $0.86 | $3.44 | In stock | Rent |
| Hyperstack | GPU cloud | 1× L40 | 48GB | On-demand | $1.00 | $1.00 | Unknown | Rent |
| Oblivus | Marketplace | 4× L40 | 192GB | Reserved | $0.74 | $2.96 | In stock | Rent |
| Vast.ai | GPU cloud | 4× L40 | 192GB | On-demand | $0.57 | $2.30 | In stock | Rent |
| Sesterce | GPU cloud | 2× L40 | 96GB | On-demand | $1.09 | $2.18 | In stock | Rent |
| Vast.ai | GPU cloud | 8× L40 | 384GB | Spot | $0.47 | $3.73 | In stock | Rent |
| Thunder Compute | Marketplace | 4× L40 | 192GB | On-demand | $0.99 | $3.96 | Unknown | Rent |
| Oblivus | Marketplace | 1× L40 | 48GB | On-demand | $1.05 | $1.05 | In stock | Rent |
| Sesterce | GPU cloud | 4× L40 | 192GB | On-demand | $1.10 | $4.40 | In stock | Rent |
| Oblivus | Marketplace | 8× L40 | 384GB | Reserved | $0.74 | $5.92 | In stock | Rent |
 CoreWeave | Enterprise | 8× L40 | 384GB | On-demand | $1.25 | $10.00 | Unknown | Rent |
| Vast.ai | GPU cloud | 8× L40 | 384GB | On-demand | $0.57 | $4.59 | In stock | Rent |
| Thunder Compute | Marketplace | 8× L40 | 384GB | On-demand | $0.99 | $7.92 | Unknown | Rent |
| Oblivus | Marketplace | 2× L40 | 96GB | On-demand | $1.05 | $2.10 | In stock | Rent |
| Oblivus | Marketplace | 4× L40 | 192GB | On-demand | $1.05 | $4.20 | In stock | Rent |
| Oblivus | Marketplace | 8× L40 | 384GB | On-demand | $1.05 | $8.40 | In stock | Rent |
Earn by hosting the L40
How this is worked out
“Renters pay” is the median live rate for this GPU on that marketplace — the median, not the floor, because the cheapest listing is one host undercutting everyone. “You earn” is that figure less the platform’s cut.
“Per day” is a ceiling: 24 hours, every hour rented, which no host achieves — scale it by the utilisation you expect. A card sitting idle earns nothing.Fees marked as published come from the platform’s terms; the rest are our estimates, labelled on each row.
Technical specifications
| FP64 | 1.4 TFLOPS |
|---|---|
| FP32 | 91 TFLOPS |
| FP16/BF16 | 181 TFLOPS |
- Memory
- 48 GB
- Memory bandwidth
- 864 GB/s
- TDP
- 300 W
- INT8
- 362 TOPS
- Die size
- 608 mm²
- Transistors
- 76.3 B
- Released
- 13 Oct 2022
| Architecture | Ada Lovelace |
|---|---|
| Memory | 48 GB GDDR6 ECC |
| Bandwidth | 864 GB/s |
| Form factor | passive dual-slot PCIe |
| Cooling | server airflow |
| Power | 300 W |
- Class
- Inference
- Cloud providers
- 9 tracked
- Price range
- $0.40 – $1.25 / GPU / hr
Other inference cards
NVIDIA A10SNAPSHOT 2026-08-23T23:42:41.203320+00:00 · BACK TO THE FULL INDEX