Legacy Cloud GPUs Pre-2020 — May 2026
Older cloud GPUs (pre-2020) still hosted by some providers. Cheapest options for inference and learning workloads.
V100 vs T4 vs P4 — top picks from this guide
|
V100
Volta · 16 GB
|
T4
Turing · 16 GB
|
P4
Pascal · 8 GB
|
|
|---|---|---|---|
| Specifications | |||
| Manufacturer | NVIDIA | NVIDIA | NVIDIA |
| Architecture | Volta | Turing | Pascal |
| VRAM | 16 GB HBM2 | 16 GB GDDR6 | 8 GB GDDR5 |
| Memory Bandwidth | 900 GB/s | 320 GB/s | 192 GB/s |
| FP16 (Tensor) | 125 TFLOPS | 65 TFLOPS | — |
| FP32 | 15.7 TFLOPS | 8.1 TFLOPS | 5.5 TFLOPS |
| TDP | 300 W | 70 W | 75 W |
| Release Year | 2017 | 2018 | 2016 |
| Segment | Data center | Data center | Data center |
| Cloud Pricing | |||
| Cheapest On-Demand | $0.13/hr | $0.08/hr | $0.16/hr |
| Providers | 1 | 1 | 1 |
Build your own GPU comparison
Select any 2 GPUs from this guide and open them side-by-side.
Tip: GPU comparisons run in pairs. Pick exactly 2 — if you skip selection, we open the top 2 from this guide.