Cheapest GPUs to run Kimi K3 2.8T locally
Kimi K3 2.8T needs about 1686 GB of VRAM at Q4 with an 8K-token context (2954 GB at Q8). A mixture of experts: 104B active. Too big for any one home machine. No PCIe setup of up to 4 cards holds it with today's listings: try the modules or 8 cards below.
The cheapest setups right now
Live on eBayNo card holds it with the live listings right now: try fewer tokens of context or a smaller quantisation.
Per question: scaled from the model's reported speed on an RTX PRO 6000 by each card's compute and memory bandwidth, within about half either way. Licenses, scores and sources: decision models.
A set of cards is priced at that many of the cheapest live listings, shipping included. Speed: an estimate from memory bandwidth for one user chatting, within a third either way.
How it works
The memory a model needs is its weights at the quantisation you pick, plus the attention cache for the context and a little for working buffers. Q4 is what most people run at home, close to full quality at a third of the size. Each card keeps about 1 GB for itself, so two 24 GB cards hold 46 GB.
Writing each token reads every active weight once, so the speed is about the memory bandwidth divided by the model's size, times what the software gets out of the card (less off CUDA). Cards sharing a model layer by layer take turns, so a set runs about as fast as one card would. Treat it as a rough figure for one person chatting; a long prompt takes longer to read in.
Power is the cards' rated power. Over about 1,440 W with the rest of the PC, a set needs more than one 15 A circuit (or a 240 V one), and older cards often need a power cable adapter, which each card's page mentions.