Skip to content

Can Mac M4 Pro 48G run Apex 35B V1.0?

Comfortable: GGUF Q5_K_M needs 28.8 GB at 4K, batch 1; headroom 7.2 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 Q3_K_M18.7ComfortableVerified file
GGUF Q4_K_M23.9ComfortableVerified file
GGUF Q5_K_M28.8ComfortableVerified file
GGUF Q8_039.6Does not fitVerified file

Context capacity · GGUF Q5_K_M

ContextTotal GBVerdict
4K28.8Comfortable
16K29.0Comfortable
32K29.4Comfortable
128K31.4Comfortable

Generation speed

GGUF Q5_K_M: 124 tok/s · Bandwidth ceiling (not measured)

Other model / GPU pairs