🔌
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 Ankh_large?
Yes — here's how fast and up to what context, with real measured numbers.
Ankh_large on the Apple M3 Max 36GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 414k | 53.7-64.7 t/s | ~1.8-2.2 (slow) |
| Q8 | yes | 443k | 107.4-129.5 t/s | ~3.6-4.3 (slow) |
| Q4 | yes | 455k | 191.7-231.2 t/s | ~6.4-7.7 (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 Ankh_large?▶
Yes. At Q4 it generates ~191.7-231.2 tok/s and fits up to 455k context; beyond that it spills to SSD swap and slows to ~6.4-7.7 tok/s.
How much context fits?▶
At Q4, up to about 455k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.