community-verified · measured pool data
Can the Ryzen AI Max 395 128GB run Qwen3-Coder-Next?
Yes — community-measured at 51.2 tok/s (5 runs at 5-bit).
Verified single-stream runs on the Ryzen AI Max 395 128GB from the localmaxxing public pool. Every number on this page is measured — nothing here is estimated.
| Quantization | Median tok/s | TTFT (ms) | Peak VRAM | Max context tested | Engines |
|---|---|---|---|---|---|
| 4-bit | 55.8 | — | 46 GB | 80,000 | llama.cpp |
| 5-bit | 51.2 | 1222.9 | 53.7 GB | 80,000 | llama.cpp |