Sunk Cost sunkcost.ai Data checked 2026-09-03

NVIDIA GeForce RTX 5090, 32GB vs NVIDIA GeForce RTX 3090, 24GB for local AI

The GeForce RTX 5090, 32GB holds 27 of the 39 open models here, and the GeForce RTX 3090, 24GB holds 24. The GeForce RTX 3090, 24GB costs $500 less, both priced as the card alone, without the PC around either. On Qwen3.8 27B, the strongest model both hold, the GeForce RTX 5090, 32GB is about 1.7× faster: 66 tok/s against 38, both estimated from memory bandwidth. The GeForce RTX 3090, 24GB pays for itself sooner, in 19 years against 25 years at 500k tokens a day. The GeForce RTX 3090, 24GB is the previous generation, so every figure here for it is priced at what it launched at rather than at a price you can pay today.

NVIDIA GeForce RTX 5090, 32GBNVIDIA GeForce RTX 3090, 24GB
Price$1,999card only$1,499card only
Memory32 GB24 GB
Usable by the GPU31 GB23 GB
Memory bandwidth1792 GB/s936 GB/s
Power under load575 W350 W
Models that fit2724
Best model it runsQwen3.8 27BQwen3.8 27B
Speed on that model66 tok/s estimated38 tok/s estimated
Pay-back on that modelPays back in 25 yearsPays back in 19 years

Run the numbers on the NVIDIA GeForce RTX 5090, 32GB · or the NVIDIA GeForce RTX 3090, 24GB

How much use it takes to pay back

Everything above is at 500k tokens a day. Pay-back moves with how much you actually run, so here are both machines on Qwen3.8 27B, the strongest model both hold, at the five levels of use the calculator names. The NVIDIA GeForce RTX 3090, 24GB pays back sooner at every level of use, so this is not a choice that turns on how hard you work it.

A day's useNVIDIA GeForce RTX 5090, 32GBNVIDIA GeForce RTX 3090, 24GB
50ka few chats a day254 years191 years
200klight assistant use64 years48 years
1Ma moderate coding-assistant day13 years9.6 years
4Mheavy coding with an agent3.2 years2.4 years
20Magents running most of the day7.6 months5.7 months

Run the NVIDIA GeForce RTX 5090, 32GB at 20M tokens a day · or the NVIDIA GeForce RTX 3090, 24GB

What the extra memory buys

The NVIDIA GeForce RTX 5090, 32GB holds 3 models the NVIDIA GeForce RTX 3090, 24GB cannot at 32k of context. The strongest of them are what the difference in memory actually buys.

ModelWeightsNeeds at 32kOn the NVIDIA GeForce RTX 5090, 32GB
Gemma 4 31B itHaiku-class20 GB26 GB48 tok/s estimated
Granite 4.2 30BHaiku-class18 GB26 GB47 tok/s estimated
Nemotron 3.5 Lightning 30B-A3BBelow every hosted tier25 GB26 GB226 tok/s estimated

The assumptions behind both columns

Both columns use the same usage: 500k tokens a day at 15:1 input to output, 32k of context, $0.17 per kWh, and today's API prices held flat. Speeds marked estimated are worked out from memory bandwidth rather than measured, and pay-back scales with them. Graphics cards are priced as the card alone, so add the PC around one before comparing it with a complete computer. Change any of it in the calculator.

More head to head: everything the NVIDIA GeForce RTX 5090, 32GB runs · everything the NVIDIA GeForce RTX 3090, 24GB runs · every other match-up · the quickest pay-back at each level of use · every model against the frontier