🔌
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 GLM-10B?
Yes — here's how fast and up to what context, with real measured numbers.
GLM-10B on the AMD Ryzen AI Max+ 395
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 572k | 8.1-9.9 t/s | ~0.3-0.4 (slow) |
| Q8 | yes | 633k | 16.1-19.7 t/s | ~0.6-0.8 (slow) |
| Q4 | yes | 660k | 28.8-35.2 t/s | ~1.1-1.4 (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 GLM-10B?▶
Yes. At Q4 it generates ~28.8-35.2 tok/s and fits up to 660k context; beyond that it spills to SSD swap and slows to ~1.1-1.4 tok/s.
How much context fits?▶
At Q4, up to about 660k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.