NVIDIA · Ada Lovelace
L40S 48 GB
A 48 GB PCIe datacenter GPU aimed at generative-AI inference, fine-tuning, rendering, and media workloads without NVLink or MIG support.
Profile checked · Primary source ↗
- Memory
- 48 GB
- Bandwidth
- 864 GB/s
- Offers
- 5
Available capacity
Observed rental offers
| GPU / provider | Billing & commitment | Per GPU | Instance total | Stock & region | Checked / actions |
|---|---|---|---|---|---|
| L40S 48 GBNebiusL40S with AMD CPU · 48 GB per GPU | HourlySpot / interruptibleMinimum term unverified | Starting priceUSD 0.74/hr per GPU | Minimum GPU count unverified | Stock not verifiedUS and Europe | Check current offer Details & price history |
| L40S 48 GBRunpodCommunity Cloud Pod · 48 GB per GPU | HourlyOn-demandMinimum term unverified | Starting priceUSD 0.79/hr per GPU | USD 0.79/hr · 1 GPU | Stock not verifiedCommunity Cloud | Check current offer Details & price history |
| L40S 48 GBRunpodSecure Cloud Pod · 48 GB per GPU | HourlyOn-demandMinimum term unverified | USD 1.09/hr per GPU | USD 1.09/hr · 1 GPU | Stock not verifiedSecure Cloud | Check current offer Details & price history |
| L40S 48 GBNebiusL40S with AMD CPU · 48 GB per GPU | HourlyOn-demandMinimum term unverified | Starting priceUSD 1.55/hr per GPU | Minimum GPU count unverified | Stock not verifiedUS and Europe | Check current offer Details & price history |
| L40S 48 GBCoreWeaveNVIDIA L40S · 48 GB per GPU | HourlyOn-demandMinimum term unverified | USD 2.25/hr per GPU | USD 18/hr · 8 GPUs | Stock not verifiedNorth America and Europe | Check current offer Details & price history |
Prices and availability can change. Confirm the current offer with the provider.
Measured performance
Inference benchmarks
| Model | GPU setup | Engine | Quant. | Workload | Output tok/s | Evidence |
|---|---|---|---|---|---|---|
| No benchmark records have been supplied yet. | ||||||