Skip to content

Can Mac M4 Pro 24G run Granite 4.2 8B?

Comfortable: GGUF Q8_0 needs 10.5 GB at 4K, batch 1; headroom 7.5 GB.

Text input, FP16 KV cache, one card. Unified-memory capacity uses the GPU guide’s usable limit; negative headroom may require host offload. Speed source settings may differ from this 4K estimate.

Measured runs

No measured run for this pair. Sizing is an estimate, not a runtime test.

Quant memory at 4K

QuantTotal GBVerdictWeight sizing
GGUF Q4_K_M6.5ComfortableVerified file
GGUF Q8_010.5ComfortableVerified file

Context capacity · GGUF Q8_0

ContextTotal GBVerdict
4K10.5Comfortable
16K12.5Comfortable
32K15.3Comfortable
128K31.8Does not fit

Generation speed

GGUF Q8_0: 20–27 tok/s · Derived range

Other model / GPU pairs