🔌
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 GePpeTto?
Yes — here's how fast and up to what context, with real measured numbers.
GePpeTto on the Apple M4 Max 36GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 468k | 1586.7-1913.3 t/s | ~29.1-35 (slow) |
| Q8 | yes | 470k | 3173.3-3826.7 t/s | ~58.1-70.1 (slow) |
| Q4 | yes | 470k | 5666.7-6833.3 t/s | ~103.8-125.2 (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 GePpeTto?▶
Yes. At Q4 it generates ~5666.7-6833.3 tok/s and fits up to 470k context; beyond that it spills to SSD swap and slows to ~103.8-125.2 tok/s.
How much context fits?▶
At Q4, up to about 470k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.