// RUN — FITS

Can the Apple M3 Max 36GB run HGRN 1B (WT 103)?

Yes — here's how fast and up to what context, with real measured numbers.

HGRN 1B (WT 103) on the Apple M3 Max 36GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 441k 102-123 t/s ~3.4-4.1 (slow)
Q8 yes 456k 204-246 t/s ~6.8-8.2 (slow)
Q4 yes 463k 364.3-439.3 t/s ~12.1-14.6 (slow)

Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.

→ All models on the Apple M3 Max 36GB → HGRN 1B (WT 103) specs → Other hardware

FAQ

Can the Apple M3 Max 36GB run HGRN 1B (WT 103)?▶

Yes. At Q4 it generates ~364.3-439.3 tok/s and fits up to 463k context; beyond that it spills to SSD swap and slows to ~12.1-14.6 tok/s.

How much context fits?▶

At Q4, up to about 463k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.