🔌
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 M1 Ultra 64GB run bge-reranker-large?
Yes — here's how fast and up to what context, with real measured numbers.
bge-reranker-large on the Apple M1 Ultra 64GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 839k | 485.7-585.7 t/s | ~6.1-7.3 (slow) |
| Q8 | yes | 847k | 971.4-1171.4 t/s | ~12.1-14.6 (slow) |
| Q4 | yes | 851k | 1734.7-2091.8 t/s | ~21.7-26.1 (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 M1 Ultra 64GB run bge-reranker-large?▶
Yes. At Q4 it generates ~1734.7-2091.8 tok/s and fits up to 851k context; beyond that it spills to SSD swap and slows to ~21.7-26.1 tok/s.
How much context fits?▶
At Q4, up to about 851k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.