Turing
NVIDIA T4
wins
2
The NVIDIA T4 is the most widely deployed inference GPU in the cloud.
Ada Lovelace
NVIDIA L4
wins
3
The NVIDIA L4 is the most power-efficient Ada Lovelace GPU at just 72W TDP.
Performance comparison
Visual bars — winner highlighted
NVIDIA T4
Metric
NVIDIA L4
FP16 TFLOPS
Bandwidth
VRAM
TDP (lower=better)
TFLOPS/watt
GB/s per watt
Full specification table
| Spec | NVIDIA T4 | NVIDIA L4 |
|---|---|---|
| Architecture | Turing | Ada Lovelace |
| VRAM | 16 GB GDDR6 | 24 GB GDDR6 |
| VRAM type | GDDR6 | GDDR6 |
| Memory bandwidth | 320 GB/s | 300 GB/s |
| FP16 TFLOPS | 65 TFLOPS | 242 TFLOPS |
| TDP | 70 W | 72 W |
| TFLOPS/watt | 0.93 T/W | 3.36 T/W |
| GB/s per watt | 4.57 GB/s/W | 4.17 GB/s/W |
| NVLink | No | No |
| Release year | 2018 | 2023 |
Live cloud pricing
On-demand $/hr across providers — updated in real time
Power efficiency analysis
TFLOPS/watt and GB/s/watt — critical for data center TCO
NVIDIA T4
NVIDIA L4
TFLOPS/watt measures compute efficiency — how much AI throughput you get per watt of power consumed. For data centers with PUE of 1.2–1.5, a 10% improvement in TFLOPS/watt translates directly to lower electricity costs and cooling requirements. GB/s/watt measures memory bandwidth efficiency, which is the binding constraint for memory-bound LLM inference workloads.
When to choose each GPU
Choose NVIDIA T4 for:
- Low-cost inference
- Edge AI
- Multi-tenant serving
Choose NVIDIA L4 for:
- Efficient inference
- Video transcoding
- Edge AI
Popular GPU comparisons