🔌
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 M2 Max 32GB run Japanese-LM-3.6B?
Yes — here's how fast and up to what context, with real measured numbers.
Japanese-LM-3.6B on the Apple M2 Max 32GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 307k | 37.8-45.6 t/s | ~0.9-1.1 (slow) |
| Q8 | yes | 362k | 75.6-91.1 t/s | ~1.9-2.3 (slow) |
| Q4 | yes | 386k | 134.9-162.7 t/s | ~3.4-4.1 (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 M2 Max 32GB run Japanese-LM-3.6B?▶
Yes. At Q4 it generates ~134.9-162.7 tok/s and fits up to 386k context; beyond that it spills to SSD swap and slows to ~3.4-4.1 tok/s.
How much context fits?▶
At Q4, up to about 386k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.