NVIDIA L4 24GB vs NVIDIA L40S 48GB

The L40S delivers 3.0x the FP8 throughput of the L4 (733 vs 243 TFLOPS dense). The L40S also carries 24 GB more memory (48 GB vs 24 GB).

L4: Ada Lovelace, 2023 L40S: Ada Lovelace, 2023
FP8 dense lead
3.0x
L40S: 733 vs 243 TFLOPS
VRAM
24 vs 48 GB
24 GB more for the L40S
Bandwidth
300 GB/s vs 864 GB/s
L40S moves data faster
TDP
72 W vs 350 W
L4 draws 278 W less

Specifications

L4
L40S
Architecture
Ada LovelaceAda Lovelace
Launch Year
20232023
Form Factor
PCIePCIe
VRAM
24 GB48 GB
Memory Bandwidth
300 GB/s864 GB/s
TDP
72 W350 W
Process Node
4nm4nm

Performance (TFLOPS)

L4
L40S
FP64
0.4 TFlops 1.4 TFlops
FP32
30.3 TFlops 91.6 TFlops
FP16
121.0 TFlops
242.0 TFLOPS sparse
362.1 TFlops
733.0 TFLOPS sparse
BF16
121.0 TFlops
242.0 TFLOPS sparse
362.1 TFlops
733.0 TFLOPS sparse
FP8
242.5 TFlops
485.0 TFLOPS sparse
733.0 TFlops
1466.0 TFLOPS sparse
INT8
242.5 TFlops
485.0 TOPS sparse
733.0 TFlops
1466.0 TOPS sparse

FLOPS by Precision

What actually differs

Both GPUs launched in 2023: the L4 on Ada Lovelace and the L40S on Ada Lovelace.

Dense throughput for the L4 against the L40S: FP64 0.4 vs 1.4 TFLOPS, FP32 30 vs 92 TFLOPS, FP16 121 vs 362 TFLOPS, FP8 243 vs 733 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 24 GB against 48 GB, fed at 300 GB/s versus 864 GB/s. More memory per GPU means larger models fit before you have to shard across cards, which often matters more than raw TFLOPS for inference.

Power budgets are 72 W for the L4 and 350 W for the L40S. At FP32 that works out to 0.42 against 0.26 TFLOPS per watt.

For LLM training and inference, weight the FP8 and FP16 rows and memory capacity most heavily. For scientific and HPC workloads, the FP64 row is the one to read.

Frequently asked questions

Is the L4 faster than the L40S?

At FP8 precision the L40S reaches 733 TFLOPS dense against 243 TFLOPS for the L4. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the L4 or the L40S?

The L40S carries 48 GB of VRAM versus 24 GB for the L4. Memory bandwidth is 300 GB/s for the L4 and 864 GB/s for the L40S.

How much power do the L4 and the L40S draw?

The L4 is rated at 72 W TDP and the L40S at 350 W. On FP32 throughput per watt, the L4 is the more efficient part.

Can I rent the L4 or the L40S in the cloud?

Yes. Live cloud listings tracked by Flopper start at $0.44 per GPU hour. The rental pricing section on this page lists current providers and rates for both GPUs.

Where to Rent

NVIDIA L4 24GB

ProviderConfigurationPrice/GPU-hrChecked
RunPod
1× L4 24GB Community $0.44 5h ago View →
RunPod
1× L4 24GB $0.49 5h ago View →
AceCloud
1× L4 24GB $0.66 5h ago View →
Scaleway
1× L4 24GB $0.91 5h ago View →
OVHcloud
2× L4 24GB $1.00 5h ago View →
All L4 listings and price history →

NVIDIA L40S 48GB

ProviderConfigurationPrice/GPU-hrChecked
DataCrunch
1× L40S 48GB Spot $0.69 5h ago View →
Nebius
1× L40S 48GB Spot $0.74 5h ago View →
RunPod
1× L40S 48GB Community $0.79 5h ago View →
Cudo Compute
1× L40S 48GB $0.87 5h ago View →
Massed Compute
1× L40S 48GB $0.88 5h ago View →
CoreWeave
8× L40S 48GB Spot $0.98 5h ago View →
RunPod
1× L40S 48GB $0.99 5h ago View →
DataCrunch
1× L40S 48GB $1.37 5h ago View →
AceCloud
1× L40S 48GB $1.50 5h ago View →
Crusoe
1× L40S 48GB $1.50 5h ago View →
Nebius
1× L40S 48GB $1.55 5h ago View →
DigitalOcean
1× L40S 48GB $1.57 5h ago View →
Vultr
1× L40S 48GB $1.67 5h ago View →
OVHcloud
4× L40S 48GB $1.80 5h ago View →
CoreWeave
8× L40S 48GB $2.25 5h ago View →
All L40S listings and price history →

Get Comparison Updates

New GPUs added weekly. Be the first to see how they compare.