ComputeAssay

Head to head · updated 2026-09-10

RTX4090 vs RTX5090: price and deliverability

RTX5090 currently rents at 0.7× the median price of RTX4090 ($0.40 vs $0.56 per GPU-hour on demand). Whether that premium is worth it depends on memory, fabric, and what you are actually running — not the sticker.

$/GPU-hour, on demand · ● A/B fabric disclosed · C · D · U undisclosed RTX4090 · Vast.ai (liv $0.37 RTX5090 · Vast.ai (liv $0.40 RTX4090 · RunPod $0.74
RTX4090 vs RTX5090, on-demand, 2026-09-10 · computeassay.com
RTX4090RTX5090
Lowest on-demand $/GPU-hr$0.37$0.40
Median on-demand$0.56$0.40
Sellers tracked21
Grade A/B offerings00
Max VRAM per GPU (tracked SKUs)24 GB32 GB

Memory is the first gate: RTX5090 carries 32 GB per GPU vs 24 GB. If your model + KV cache does not fit, price is irrelevant — the smaller card is not a substitute at any discount (ASSAY-1 §5).

Same-name capacity is not one good: within each column, prices spread by multiples depending on fabric. Check the per-GPU pages for every listing with its published interconnect, and the substitutability rules for when one can genuinely replace the other.

These ratios move weekly

The Delivered Compute Report tracks the spread — free, every row sourced.

Subscribe free