🔌
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 BiomedGPT (182M)?
Yes — here's how fast and up to what context, with real measured numbers.
BiomedGPT (182M) on the Apple M3 Max 36GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 466k | 560.4-675.8 t/s | ~18.7-22.5 (slow) |
| Q8 | yes | 469k | 1120.9-1351.6 t/s | ~37.4-45.1 (slow) |
| Q4 | yes | 470k | 2001.6-2413.7 t/s | ~66.7-80.5 (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 M3 Max 36GB run BiomedGPT (182M)?▶
Yes. At Q4 it generates ~2001.6-2413.7 tok/s and fits up to 470k context; beyond that it spills to SSD swap and slows to ~66.7-80.5 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.