🔌
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 Granite 3.1 2B?
Yes — here's how fast and up to what context, with real measured numbers.
Granite 3.1 2B on the AMD Ryzen AI Max+ 395
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1659k | 32.3-39.4 t/s | ~1.3-1.5 (slow) |
| Q8 | yes | 1697k | 64.5-78.8 t/s | ~2.5-3.1 (slow) |
| Q4 | yes | 1714k | 115.2-140.8 t/s | ~4.5-5.5 (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 Granite 3.1 2B?▶
Yes. At Q4 it generates ~115.2-140.8 tok/s and fits up to 1714k context; beyond that it spills to SSD swap and slows to ~4.5-5.5 tok/s.
How much context fits?▶
At Q4, up to about 1714k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.