// RUN — FITS

Can the Apple M2 Max 96GB run CausalLM 34B β?

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

CausalLM 34B β on the Apple M2 Max 96GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 61k 4-4.8 t/s ~0.1-0.1 (slow)
Q8 yes 193k 7.9-9.5 t/s ~0.2-0.2 (slow)
Q4 yes 250k 14.1-17 t/s ~0.4-0.4 (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 M2 Max 96GB → CausalLM 34B β specs → Other hardware

FAQ

Can the Apple M2 Max 96GB run CausalLM 34B β?▶

Yes. At Q4 it generates ~14.1-17 tok/s and fits up to 250k context; beyond that it spills to SSD swap and slows to ~0.4-0.4 tok/s.

How much context fits?▶

At Q4, up to about 250k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.