🔌
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 HiFi - NN?
Yes — here's how fast and up to what context, with real measured numbers.
HiFi - NN on the Apple M3 Max 36GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 471k | 34000-41000 t/s | ~1133.3-1366.7 (slow) |
| Q8 | yes | 471k | 68000-82000 t/s | ~2266.7-2733.3 (slow) |
| Q4 | yes | 471k | 121428.6-146428.6 t/s | ~4047.6-4881 (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 HiFi - NN?▶
Yes. At Q4 it generates ~121428.6-146428.6 tok/s and fits up to 471k context; beyond that it spills to SSD swap and slows to ~4047.6-4881 tok/s.
How much context fits?▶
At Q4, up to about 471k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.