Skip to content

Can Mac M4 24G run Clover-35B-A3B-Preview?

Tight: GGUF Q3_K_M needs 18.2 GB at 4K, batch 1; headroom -0.2 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_M18.2TightVerified file
GGUF Q4_020.9Does not fitVerified file
GGUF Q4_K_M23.4Does not fitVerified file
GGUF Q5_K_M28.3Does not fitVerified file
GGUF Q8_038.6Does not fitVerified file

Context capacity · GGUF Q3_K_M

ContextTotal GBVerdict
4K18.2Tight
16K18.4Tight
32K18.8Tight
128K20.9Does not fit

Generation speed

GGUF Q3_K_M: 84 tok/s · Bandwidth ceiling (not measured)

Other model / GPU pairs