🔌
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 M3 Max 36GB run RWKV-5 (Eagle) 7B?
Yes — here's how fast and up to what context, with real measured numbers.
RWKV-5 (Eagle) 7B on the Apple M3 Max 36GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 121k | 13.6-16.4 t/s | ~0.5-0.5 (slow) |
| Q8 | yes | 178k | 27.1-32.7 t/s | ~0.9-1.1 (slow) |
| Q4 | yes | 204k | 48.4-58.4 t/s | ~1.6-1.9 (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 M3 Max 36GB run RWKV-5 (Eagle) 7B?▶
Yes. At Q4 it generates ~48.4-58.4 tok/s and fits up to 204k context; beyond that it spills to SSD swap and slows to ~1.6-1.9 tok/s.
How much context fits?▶
At Q4, up to about 204k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.