AI hardware
NVIDIA L4
FP16 throughput
1.2×10¹⁴ FLOP/s
Memory
22.4 GB
TDP
72 W
NVIDIA L4 is an AI accelerator designed by NVIDIA (GPU), released in March 2023. It is built on a 5 nm process at TSMC, packing 35.8B transistors.
It delivers 1.2×10¹⁴ FLOP/s of tensor FP16/BF16 throughput. It carries 22.4 GB of memory at 279 GB/s of bandwidth. Its thermal design power is 72 W.
Full record
- Manufacturer
- NVIDIA
- Type
- GPU
- Release date
- Mar 21, 2023
- Tensor FP16/BF16
- 1.2×10¹⁴ FLOP/s
- FP32
- 3×10¹³ FLOP/s
- FP64
- 4.7×10¹¹ FLOP/s
- INT8
- 2.4×10¹⁴ OP/s
- Memory
- 22.4 GB
- Memory bandwidth
- 279 GB
- TDP
- 72 W
- Process node
- 5 nm
- Transistors (millions)
- 35,800
- Die size (mm2)
- 294
- Foundry
- TSMC
- Tensor cores
- 240
More from NVIDIA
SourceEpoch AI, 'Machine Learning Hardware'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/data/machine-learning-hardware. Licensed under CC BY 4.0.
← All ai hardware