🔌
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 M2 Ultra 64GB run OpenELM-450M?
Yes — here's how fast and up to what context, with real measured numbers.
OpenELM-450M on the Apple M2 Ultra 64GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 842k | 604.4-728.9 t/s | ~7.6-9.1 (slow) |
| Q8 | yes | 849k | 1208.9-1457.8 t/s | ~15.1-18.2 (slow) |
| Q4 | yes | 852k | 2158.7-2603.2 t/s | ~27-32.5 (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 M2 Ultra 64GB run OpenELM-450M?▶
Yes. At Q4 it generates ~2158.7-2603.2 tok/s and fits up to 852k context; beyond that it spills to SSD swap and slows to ~27-32.5 tok/s.
How much context fits?▶
At Q4, up to about 852k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.