A100-PCIE vs A100-PCIE-80GB
A100-PCIE is the cheaper of the two: $0.74/hr at Vast.ai, against $1.35/hr for A100-PCIE-80GB at Shadeform — 1.8× the price.
Cheapest on each
Aligned facts
Same sheet, same order, both cards — including the gaps.
| Attribute | A100-PCIE | A100-PCIE-80GB |
|---|---|---|
| VRAM | ○ no answer | 80 GB |
| Bandwidth | ○ no answer | 1.935 TB/s |
| FP16 | ○ no answer | 312 TFLOPS |
| FP8 | ○ no answer | ○ no answer |
| TDP | ○ no answer | 300 W |
| NVLink | ○ no answer | NVIDIA ® NVLink ® Bridge for 2 GPUs: 600 GB/s ** PCIe Gen4: 64 GB/s |
| Architecture | Ampere | Ampere |
| Released | ○ no answer | ○ no answer |
A blank is not a zero. Where one card states a figure and the other states none, neither wins the row — the second number is unknown, and ranking an unknown as nothing is how a comparison starts to lie.
Every seller, both cards
One row per seller, cheapest tier, on-demand list prices only. Sellers of both cards lead, cheapest combined rate first.
| Seller | A100-PCIE | A100-PCIE-80GB |
|---|---|---|
| Vast.ai | $0.742026-10-27 | ○ no answer |
| RunPod | $1.792026-10-27 | ○ no answer |
| Shadeform | ○ no answer | $1.352026-10-26 |
| Oblivus | ○ no answer | $1.472026-10-26 |
What this page will not tell you
On-demand list rates only · spot and committed terms priced separately and not mixed in here · no performance-per-dollar figure appears on this page, because it needs a throughput benchmark we do not hold and dividing a price by a vendor’s own TFLOPS claim produces a number that looks authoritative and means nothing