L4 vs T4

Compare NVIDIA L4 and NVIDIA T4 specs, performance, and cloud pricing

L4

24GB

From $0.350/hr

T4

16GB

From $0.220/hr

Architecture

Ada Lovelace

vs Turing

FP16 Gap

1.9x

L4 leads

SpecificationL4T4
VRAM24 GB16 GB
VRAM TypeGDDR6GDDR6
FP16 TFLOPS242 TFLOPS130 TFLOPS
FP8 TFLOPS485 TFLOPSN/A
Memory Bandwidth300 GB/s320 GB/s
TDP72W70W
InterconnectPCIe Gen4PCIe Gen3
ArchitectureAda LovelaceTuring

Price Comparison

MetricL4T4
Cheapest On-Demand$0.350/hr$0.220/hr
Cheapest Spot$0.210/hr$0.120/hr
Providers Available35

Verdict

Best for Training

NVIDIA L4

242 TFLOPS FP16 with 24GB VRAM

Best Value

NVIDIA L4

691 TFLOPS per $/hr

Best for Inference

NVIDIA L4

485 TFLOPS FP8/FP16

Use-Case Recommendations

Large-Scale Training

Training LLMs and large multi-modal models

Winner

L4

242 TFLOPS FP16 with 24GB GDDR6 provides the best training throughput.

Inference at Scale

Deploying models in production for real-time inference

Winner

L4

485 TFLOPS FP8/FP16 gives superior inference throughput.

Budget-Conscious Workloads

Getting the best performance per dollar

Winner

L4

Starting at $0.350/hr delivers the best TFLOPS per dollar.

Learn More