Llama on Apple M3
Can I run Llama 3.1 70B on an Mac Studio M3 Ultra 256GB?
Yes. Llama 3.1 70B at Q4_K_M uses 43.1 GB of the 192.0 GB available on a Mac Studio M3 Ultra 256GB, and runs at about 11.4 tokens per second. You can push the context to 128K.
Runs comfortably
43.1 GB of 192.0 GB · 22%0192 GB
Weights
40.0 GB
KV cache
2.5 GB
Runtime overhead
0.6 GB
Llama 3.1 70B at Q4_K_M leaves 148.9 GB spare on a Mac Studio M3 Ultra 256GB. There is room to raise the context length or move up a quantisation level.
Generation11.4tok/s
Prompt processing170tok/s
Max context128Ktokens
KV per 1K tokens0GB
Every quantisation of Llama 3.1 70B on a Mac Studio M3 Ultra 256GB
Highlighted row is the highest quality that still fits at 8K context.
| Quantisation | bpw | Weights | Total | Verdict | Max ctx | tok/s |
|---|---|---|---|---|---|---|
| FP16 / BF16 | 16.00 | 131.4 GB | 134.5 GB | Runs comfortably | 128K | 3.56 |
| INT8 / W8A8 | 8.50 | 69.3 GB | 72.4 GB | Runs comfortably | 128K | 6.68 |
| Q8_0 (GGUF) | 8.50 | 69.3 GB | 72.4 GB | Runs comfortably | 128K | 6.68 |
| FP8 (E4M3) | 8.00 | 65.7 GB | 68.8 GB | Runs comfortably | 128K | 7.04 |
| Q6_K | 6.56 | 53.9 GB | 57.0 GB | Runs comfortably | 128K | 8.55 |
| Q5_K_M | 5.67 | 46.6 GB | 49.7 GB | Runs comfortably | 128K | 9.85 |
| Q5_K_S | 5.52 | 45.3 GB | 48.4 GB | Runs comfortably | 128K | 10.1 |
| Q4_K_M | 4.85 | 40.0 GB | 43.1 GB | Runs comfortably | 128K | 11.4 |
| Q4_K_S | 4.58 | 37.8 GB | 40.9 GB | Runs comfortably | 128K | 12.0 |
| AWQ 4-bit | 4.25 | 37.8 GB | 40.9 GB | Runs comfortably | 128K | 12.1 |
| GPTQ 4-bit | 4.25 | 37.8 GB | 40.9 GB | Runs comfortably | 128K | 12.1 |
| MXFP4 | 4.25 | 37.8 GB | 40.9 GB | Runs comfortably | 128K | 12.1 |
| Q4_0 | 4.55 | 37.6 GB | 40.7 GB | Runs comfortably | 128K | 12.1 |
| IQ4_XS | 4.25 | 35.2 GB | 38.3 GB | Runs comfortably | 128K | 12.9 |
| Q3_K_M | 3.91 | 32.5 GB | 35.6 GB | Runs comfortably | 128K | 13.9 |
| IQ3_M | 3.70 | 30.8 GB | 33.9 GB | Runs comfortably | 128K | 14.6 |
| IQ3_XXS | 3.06 | 25.7 GB | 28.8 GB | Runs comfortably | 128K | 17.4 |
| Q2_K | 2.63 | 22.3 GB | 25.4 GB | Runs comfortably | 128K | 19.9 |
| IQ2_XXS | 2.06 | 17.8 GB | 20.9 GB | Runs comfortably | 128K | 24.6 |
| IQ1_M | 1.75 | 15.3 GB | 18.4 GB | Runs comfortably | 128K | 28.2 |