🔌
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 64GB run MatterSim (M3GNet - MatterSim-v1.0.0-5M)?
Yes — here's how fast and up to what context, with real measured numbers.
MatterSim (M3GNet - MatterSim-v1.0.0-5M) on the Apple M4 Max 64GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 856k | 41253.3-49746.7 t/s | ~755.6-911.1 (slow) |
| Q8 | yes | 856k | 82506.7-99493.3 t/s | ~1511.1-1822.2 (slow) |
| Q4 | yes | 856k | 147333.3-177666.7 t/s | ~2698.4-3254 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.
→ All models on the Apple M4 Max 64GB
→ MatterSim (M3GNet - MatterSim-v1.0.0-5M) specs
→ Other hardware
FAQ
Can the Apple M4 Max 64GB run MatterSim (M3GNet - MatterSim-v1.0.0-5M)?▶
Yes. At Q4 it generates ~147333.3-177666.7 tok/s and fits up to 856k context; beyond that it spills to SSD swap and slows to ~2698.4-3254 tok/s.
How much context fits?▶
At Q4, up to about 856k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.