Mac Studio, M5 Max, 128GB
512GB of storage, 614GB/s of memory bandwidth, and 120GB of unified memory left for a model once macOS has taken its share. It holds 8 of the 11 models we track.
Offers
checked 9 hours ago- Applelist pricein stockchecked 9 hours ago$5,099
Timing
Collecting price history — 1 of 90 days. No timing verdict yet.
Models it runs — 8K context
- Won't fit— headroom
- Won't fit— headroom
- Won't fit— headroom
- Qwen3-235B-A22BQ3_K_MTight2GB headroomest. 28.6 tok/s
- Fits15GB headroomest. 4.0 tok/s
- gpt-oss-120bQ6_KFits22GB headroomest. 73.4 tok/s
- Tight1GB headroomest. 17.0 tok/s
- Llama-3.3-70BQ8_0Fits42GB headroomest. 5.4 tok/s
- Qwen3-32BBF16Fits53GB headroomest. 6.2 tok/s
- Gemma-3-27BBF16Fits63GB headroomest. 7.4 tok/s
- gpt-oss-20bBF16Fits76GB headroomest. 42.6 tok/s
| Model | Best quant | Fit | Headroom | Est. speed |
|---|---|---|---|---|
| Llama-4-Behemoth | — | Won't fit | — | — |
| DeepSeek-V3 | — | Won't fit | — | — |
| Llama-4-Maverick-17B-128E | — | Won't fit | — | — |
| Qwen3-235B-A22B | Q3_K_M | Tight | 2GB | 28.6 tok/s |
| Mistral-Large-2411 | Q6_K | Fits | 15GB | 4.0 tok/s |
| gpt-oss-120b | Q6_K | Fits | 22GB | 73.4 tok/s |
| Llama-4-Scout-17B-16E | Q8_0 | Tight | 1GB | 17.0 tok/s |
| Llama-3.3-70B | Q8_0 | Fits | 42GB | 5.4 tok/s |
| Qwen3-32B | BF16 | Fits | 53GB | 6.2 tok/s |
| Gemma-3-27B | BF16 | Fits | 63GB | 7.4 tok/s |
| gpt-oss-20b | BF16 | Fits | 76GB | 42.6 tok/s |
Best quant is the highest-quality one that still loads with headroom. Every speed is an estimate; the formula is here.