🔌
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 AMD Ryzen AI Max+ 395 run Adaptive Inputs + LayerDrop?
Yes — here's how fast and up to what context, with real measured numbers.
Adaptive Inputs + LayerDrop on the AMD Ryzen AI Max+ 395
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1722k | 190.6-233 t/s | ~7.4-9.1 (slow) |
| Q8 | yes | 1728k | 381.3-466 t/s | ~14.9-18.2 (slow) |
| Q4 | yes | 1731k | 680.9-832.2 t/s | ~26.6-32.5 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.
FAQ
Can the AMD Ryzen AI Max+ 395 run Adaptive Inputs + LayerDrop?▶
Yes. At Q4 it generates ~680.9-832.2 tok/s and fits up to 1731k context; beyond that it spills to SSD swap and slows to ~26.6-32.5 tok/s.
How much context fits?▶
At Q4, up to about 1731k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.