🔌
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 M1 Ultra 64GB run Llama 3.2 3B?
Yes — here's how fast and up to what context, with real measured numbers.
Llama 3.2 3B on the Apple M1 Ultra 64GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 758k | 84.7-102.2 t/s | ~1.1-1.3 (slow) |
| Q8 | yes | 807k | 169.5-204.4 t/s | ~2.1-2.6 (slow) |
| Q4 | yes | 829k | 302.6-364.9 t/s | ~3.8-4.6 (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 M1 Ultra 64GB run Llama 3.2 3B?▶
Yes. At Q4 it generates ~302.6-364.9 tok/s and fits up to 829k context; beyond that it spills to SSD swap and slows to ~3.8-4.6 tok/s.
How much context fits?▶
At Q4, up to about 829k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.