🔌
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 Pythia-1.4b?
Yes — here's how fast and up to what context, with real measured numbers.
Pythia-1.4b on the Apple M3 Max 36GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 429k | 72.9-87.9 t/s | ~2.4-2.9 (slow) |
| Q8 | yes | 450k | 145.7-175.7 t/s | ~4.9-5.9 (slow) |
| Q4 | yes | 460k | 260.2-313.8 t/s | ~8.7-10.5 (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 Pythia-1.4b?▶
Yes. At Q4 it generates ~260.2-313.8 tok/s and fits up to 460k context; beyond that it spills to SSD swap and slows to ~8.7-10.5 tok/s.
How much context fits?▶
At Q4, up to about 460k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.