Skip to content

Can Mac M1 8G run LFM2.5 8B-A1B?

Tight: GGUF Q4_K_M needs 5.4 GB at 4K, batch 1; headroom 0.6 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_M5.4TightVerified file
GGUF Q5_K_M6.4Does not fitVerified file
GGUF Q8_09.5Does not fitVerified file

Context capacity · GGUF Q4_K_M

ContextTotal GBVerdict
4K5.4Tight
16K5.6Tight
32K5.8Tight

Generation speed

GGUF Q4_K_M: 78 tok/s · Bandwidth ceiling (not measured)

Other model / GPU pairs