🔌
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 iGPT-L?
Yes — here's how fast and up to what context, with real measured numbers.
iGPT-L on the Apple M2 Ultra 64GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 814k | 199.7-240.8 t/s | ~2.5-3 (slow) |
| Q8 | yes | 835k | 399.4-481.6 t/s | ~5-6 (slow) |
| Q4 | yes | 844k | 713.2-860.1 t/s | ~8.9-10.8 (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 iGPT-L?▶
Yes. At Q4 it generates ~713.2-860.1 tok/s and fits up to 844k context; beyond that it spills to SSD swap and slows to ~8.9-10.8 tok/s.
How much context fits?▶
At Q4, up to about 844k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.