// RUN — FITS

Can the Apple M1 Max 32GB run Nucleotide Transformer?

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

Nucleotide Transformer on the Apple M1 Max 32GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 340k 54.4-65.6 t/s ~1.4-1.6 (slow)
Q8 yes 378k 108.8-131.2 t/s ~2.7-3.3 (slow)
Q4 yes 395k 194.3-234.3 t/s ~4.9-5.9 (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 M1 Max 32GB → Nucleotide Transformer specs → Other hardware

FAQ

Can the Apple M1 Max 32GB run Nucleotide Transformer?▶

Yes. At Q4 it generates ~194.3-234.3 tok/s and fits up to 395k context; beyond that it spills to SSD swap and slows to ~4.9-5.9 tok/s.

How much context fits?▶

At Q4, up to about 395k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.