Home / Will it run? / Mistral Small 3.2 24B Instruct (2506)
Mistral AI · Mistral Small 3
Will Mistral Small 3.2 24B Instruct (2506) run on my GPU?
24B parameters, 131,072-token context, Apache-2.0. At Q4_K_M the weights come to about 13.4 GiB; add 1.5 GiB for the runtime and you need roughly 14.9 GiB of VRAM before context.
Weights by quant
GiBQ4_K_M · the everyday quant13.4 GiB
Q8_0 · near-lossless23.7 GiB
FP16 · full weights44.7 GiB
Runtime reserve1.5 GiB
Card by card
11 of 17 fit at Q4Headroom is what remains for the KV cache after weights and the 1.5 GiB reserve. How far it stretches depends on the architecture: grouped-query models are frugal, older dense models are not. If the table says tight, plan on a shorter context or a lower quant. Parameter count from the model's own card.