🔌
New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
// RUN — FITS
Can the Apple M3 Max 36GB run Engine-Base (NE)?
Yes — here's how fast and up to what context, with real measured numbers.
Engine-Base (NE) on the Apple M3 Max 36GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 468k | 822.6-991.9 t/s | ~27.4-33.1 (slow) |
| Q8 | yes | 470k | 1645.2-1983.9 t/s | ~54.8-66.1 (slow) |
| Q4 | yes | 470k | 2937.8-3542.6 t/s | ~97.9-118.1 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.
FAQ
Can the Apple M3 Max 36GB run Engine-Base (NE)?▶
Yes. At Q4 it generates ~2937.8-3542.6 tok/s and fits up to 470k context; beyond that it spills to SSD swap and slows to ~97.9-118.1 tok/s.
How much context fits?▶
At Q4, up to about 470k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.