RTX-4090-24GB vs RTX-A4000-16GB
RTX-A4000-16GB is the cheaper of the two: $0.1/hr at TensorDock, against $0.16/hr for RTX-4090-24GB at Salad — 1.6× the price.
Cheapest on each
Aligned facts
Same sheet, same order, both cards — including the gaps.
| Attribute | RTX-4090-24GB | RTX-A4000-16GB |
|---|---|---|
| VRAM | 24 GB▲ | 16 GB |
| Bandwidth | 1.008 TB/s | ○ no answer |
| FP16 | 165 TFLOPS | ○ no answer |
| FP8 | 330 TFLOPS | ○ no answer |
| TDP | 450 W▲ | 140 W |
| NVLink | PCIe 4 (no NVLink) | ○ no answer |
| Architecture | Ada Lovelace | ○ no answer |
| Released | 2022 | ○ 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 | RTX-4090-24GB | RTX-A4000-16GB |
|---|---|---|
| TensorDock | $0.352026-10-26 | $0.102026-10-26 |
| Lium | $0.652026-10-27JSON | $0.122026-10-27JSON |
| RunPod | $0.892026-10-27 | $0.252026-10-27 |
| Runcrate | $0.922026-10-26 | $1.042026-10-26 |
| Salad | $0.162026-10-26 | ○ no answer |
| Theta EdgeCloud | $0.262026-10-26 | ○ no answer |
| Packet·ai | $0.392026-10-27 | ○ no answer |
| Shadeform | $0.402026-10-26 | ○ no answer |
| Beam | $0.442026-10-27 | ○ no answer |
| GPUhub | $0.462026-10-26 | ○ no answer |
| Vast.ai | $0.472026-10-27 | ○ no answer |
| Oblivus | $0.642026-10-26 | ○ no answer |
| Spheron | $0.722026-10-26 | ○ no answer |
| Hyperstack | ○ no answer | $0.152026-10-26 |
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