🔌
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 36GB run MSA Transformer?
Yes — here's how fast and up to what context, with real measured numbers.
MSA Transformer on the Apple M4 Max 36GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 468k | 1856.4-2238.6 t/s | ~34-41 (slow) |
| Q8 | yes | 470k | 3712.8-4477.2 t/s | ~68-82 (slow) |
| Q4 | yes | 471k | 6630-7995 t/s | ~121.4-146.4 (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 36GB run MSA Transformer?▶
Yes. At Q4 it generates ~6630-7995 tok/s and fits up to 471k context; beyond that it spills to SSD swap and slows to ~121.4-146.4 tok/s.
How much context fits?▶
At Q4, up to about 471k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.