🔌
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 Max 32GB run DiffDock?
Yes — here's how fast and up to what context, with real measured numbers.
DiffDock on the Apple M1 Max 32GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 416k | 6719.4-8102.8 t/s | ~168-202.6 (slow) |
| Q8 | yes | 416k | 13438.7-16205.5 t/s | ~336-405.1 (slow) |
| Q4 | yes | 416k | 23997.7-28938.5 t/s | ~599.9-723.5 (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 Max 32GB run DiffDock?▶
Yes. At Q4 it generates ~23997.7-28938.5 tok/s and fits up to 416k context; beyond that it spills to SSD swap and slows to ~599.9-723.5 tok/s.
How much context fits?▶
At Q4, up to about 416k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.