🔌
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 gpt-oss-20b?
Yes — here's how fast and up to what context, with real measured numbers.
gpt-oss-20b on the Apple M1 Ultra 64GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 54k | 13-15.7 t/s | ~0.2-0.2 (slow) |
| Q8 | yes | 134k | 26-31.4 t/s | ~0.3-0.4 (slow) |
| Q4 | yes | 169k | 46.5-56 t/s | ~0.6-0.7 (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 gpt-oss-20b?▶
Yes. At Q4 it generates ~46.5-56 tok/s and fits up to 169k context; beyond that it spills to SSD swap and slows to ~0.6-0.7 tok/s.
How much context fits?▶
At Q4, up to about 169k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.