Skip to content

Can Mac M4 Max 36G run MiniCPM-V 4.7 35B-A3B?

Comfortable: GGUF Q4_K_M needs 22.2 GB at 4K, batch 1; headroom 4.8 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_S20.9ComfortableVerified file
GGUF Q4_K_M22.2ComfortableVerified file
GGUF Q5_K_M25.9TightVerified file
GGUF Q8_038.6Does not fitVerified file

Context capacity · GGUF Q4_K_M

ContextTotal GBVerdict
4K22.2Comfortable
16K22.5Comfortable
32K22.8Comfortable
128K24.9Tight

Generation speed

GGUF Q4_K_M: 236 tok/s · Bandwidth ceiling (not measured)

Other model / GPU pairs