Skip to content

Can Mac M4 Max 36G run K2 Horizon MoVA 36B-A4B?

Tight: GGUF Q4_K_M needs 24.2 GB at 4K, batch 1; headroom 2.8 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_M24.2TightVerified file
GGUF Q5_K_M28.5Does not fitVerified file
GGUF Q6_K33.0Does not fitVerified file
GGUF Q8_042.5Does not fitVerified file

Context capacity · GGUF Q4_K_M

ContextTotal GBVerdict
4K24.2Tight
16K26.7Tight
32K30.0Does not fit
128K49.8Does not fit

Generation speed

GGUF Q4_K_M: 181 tok/s · Bandwidth ceiling (not measured)

Other model / GPU pairs