// RUN — FITS

Can the Apple M3 Max 36GB run DeepSeek-VL-1.3B?

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

DeepSeek-VL-1.3B on the Apple M3 Max 36GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 432k 78.5-94.6 t/s ~2.6-3.2 (slow)
Q8 yes 452k 156.9-189.2 t/s ~5.2-6.3 (slow)
Q4 yes 460k 280.2-337.9 t/s ~9.3-11.3 (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 → DeepSeek-VL-1.3B specs → Other hardware

FAQ

Can the Apple M3 Max 36GB run DeepSeek-VL-1.3B?▶

Yes. At Q4 it generates ~280.2-337.9 tok/s and fits up to 460k context; beyond that it spills to SSD swap and slows to ~9.3-11.3 tok/s.

How much context fits?▶

At Q4, up to about 460k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.