GPU comparison
NVIDIA A100 40GB vs NVIDIA RTX 3090.
Specs, performance and live cloud availability, side by side.
The short answer
Choose NVIDIA A100 40GB if you need 40GB+ on a single GPU (16GB more than NVIDIA RTX 3090).
Choose NVIDIA A100 40GB if you need FP16 tensor throughput (339% faster than NVIDIA RTX 3090).
Compare another pair
vs
Ampere
NVIDIA A100 40GB
from$1.99/hr
Available now · 5 regions- Memory
- 40 GB HBM2
- FP16 tensor
- 312 TFLOPS
- Bandwidth
- 1.6 TB/s
Ampere
NVIDIA RTX 3090
Not on demand yet- Memory
- 24 GB GDDR6X
- FP16 tensor
- 71 TFLOPS
- Bandwidth
- 936 GB/s
At a glance
Spec by spec. The longer bar is the better number.
Full specs
The complete spec sheet.
| Spec | NVIDIA A100 40GB | NVIDIA RTX 3090 |
|---|---|---|
| Architecture | Ampere | Ampere |
| Memory | 40 GB | 24 GB |
| Memory type | HBM2 | GDDR6X |
| Memory bandwidth | 1.6 TB/s | 936 GB/s |
| FP32 | 19.5 TFLOPS | 35.6 TFLOPS |
| FP16 tensor | 312 TFLOPS | 71 TFLOPS |
| INT8 | 624 TOPS | 284 TOPS |
| TDP | up to 400 W | 350 W |
| GPUs per node | 8 | 2 |
| Hourly price (live) | from $1.99/hr | Not on demand yet |
Questions
A100 40GB or RTX 3090. What buyers ask first.
Other comparisons