GPUs for fine-tuning with LoRA or QLoRA
One card, adapter weights only, and the base model quantised.
The rule this page selects on
24GB or more of memory, cheapest on-demand rate first.
LoRA trains a small adapter and leaves the base model frozen, and QLoRA quantises that base to 4-bit. A 7B model fits in 24GB that way, which is why this floor is a quarter of the training one rather than the same number.
20 priced parts carry no memory figure on record and are excluded rather than assumed to qualify. Unknown is not a yes.
Every part that meets it
On-demand list rates, read from each provider’s own page within 21 days · open any part for every provider selling it
What this ranking does not know
It orders by hourly rate, which is not the same as cost. A card at half the price that takes three times as long is dearer, and we hold no benchmark that would tell you which — so no page here claims a cost per unit of work. Nor does it know about FP64 throughput, interconnect bandwidth or encoder blocks. If those decide your choice, this page has narrowed the field and not made the decision.