Skip to content

Can Mac M3 8G run TwIL 8B balanced?

Tight: GGUF Q4_K_M needs 6.3 GB at 4K, batch 1; headroom -0.3 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_M6.3TightVerified file

Context capacity · GGUF Q4_K_M

ContextTotal GBVerdict
4K6.3Tight
16K8.3Does not fit
32K11.1Does not fit
128K27.6Does not fit

Generation speed

GGUF Q4_K_M: 15–20 tok/s · Derived range

Other model / GPU pairs