作者: Jakub Rusinowski · 最后更新: 2026年7月10日
Yes — the Llama 3.2 Family family runs on the AMD Ryzen AI Max+ 395.
在亚马逊查看价格 — GMKtec EVO-X2 (Ryzen AI Max+ 395, 128GB)
| VRAM | 96 GB |
| Memory Bandwidth | 256 GB/s |
| Llama 3.2 90B Vision Instruct | Q4_K_M · 54 GB · ~5 tok/s (est.) |
| Llama 3.2 11B Vision Instruct | Q4_K_M · 7.8 GB · ~33 tok/s (est.) |
| Llama 3.2 3B Instruct | Q4_K_M · 2.2 GB · ~116 tok/s (est.) |
| Llama 3.2 1B Instruct | Q4_K_M · 0.8 GB · ~320 tok/s (est.) |
Yes — the Llama 3.2 Family family runs on the AMD Ryzen AI Max+ 395.
Llama 3.2 90B Vision Instruct at Q4_K_M quantization (54 GB), estimated ~5 tokens/sec.
← Can I Run It? | Llama 3.2 Family Model Page | AMD Ryzen AI Max+ 395 GPU Page | Check Your Hardware