🔌
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 M4 Max 48GB run Qarasu-14B?
Yes — here's how fast and up to what context, with real measured numbers.
Qarasu-14B on the Apple M4 Max 48GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 84k | 13.3-16 t/s | ~0.2-0.3 (slow) |
| Q8 | yes | 169k | 26.5-32 t/s | ~0.5-0.6 (slow) |
| Q4 | yes | 207k | 47.4-57.1 t/s | ~0.9-1 (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 M4 Max 48GB run Qarasu-14B?▶
Yes. At Q4 it generates ~47.4-57.1 tok/s and fits up to 207k context; beyond that it spills to SSD swap and slows to ~0.9-1 tok/s.
How much context fits?▶
At Q4, up to about 207k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.