作者: Jakub Rusinowski · 最后更新: 2026年7月10日
Yes — the GPT-oss 120B family runs on the Apple M2 Max.
在亚马逊查看价格 — Apple Mac Studio M2 Max
| VRAM | 96 GB unified memory |
| Memory Bandwidth | 400 GB/s |
| GPT-oss 120B | Q4_K_M · 65 GB · ~6 tok/s (est.) |
Yes — the GPT-oss 120B family runs on the Apple M2 Max.
GPT-oss 120B at Q4_K_M quantization (65 GB), estimated ~6 tokens/sec.
← Can I Run It? | GPT-oss 120B Model Page | Apple M2 Max GPU Page | Check Your Hardware