🔌
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 Max 32GB run CodeT5-base?
Yes — here's how fast and up to what context, with real measured numbers.
CodeT5-base on the Apple M1 Max 32GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 410k | 618.2-745.5 t/s | ~15.5-18.6 (slow) |
| Q8 | yes | 413k | 1236.4-1490.9 t/s | ~30.9-37.3 (slow) |
| Q4 | yes | 415k | 2207.8-2662.3 t/s | ~55.2-66.6 (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 Max 32GB run CodeT5-base?▶
Yes. At Q4 it generates ~2207.8-2662.3 tok/s and fits up to 415k context; beyond that it spills to SSD swap and slows to ~55.2-66.6 tok/s.
How much context fits?▶
At Q4, up to about 415k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.