Llama on AMD Strix Halo
Can I run Llama 3.1 70B on an Ryzen AI Max+ 395 128GB?
Yes. Llama 3.1 70B at Q4_K_M uses 43.1 GB of the 96.0 GB available on a Ryzen AI Max+ 395 128GB, and runs at about 3.26 tokens per second. You can push the context to 128K.
Runs comfortably
43.1 GB of 96.0 GB · 45%096 GB
Weights
40.0 GB
KV cache
2.5 GB
Runtime overhead
0.6 GB
Llama 3.1 70B at Q4_K_M leaves 52.9 GB spare on a Ryzen AI Max+ 395 128GB. There is room to raise the context length or move up a quantisation level.
Generation3.26tok/s
Prompt processing149tok/s
Max context128Ktokens
KV per 1K tokens0GB
Every quantisation of Llama 3.1 70B on a Ryzen AI Max+ 395 128GB
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 | Won't fit | — | 0.53 |
| INT8 / W8A8 | 8.50 | 69.3 GB | 72.4 GB | Runs comfortably | 83K | 1.90 |
| Q8_0 (GGUF) | 8.50 | 69.3 GB | 72.4 GB | Runs comfortably | 83K | 1.90 |
| FP8 (E4M3) | 8.00 | 65.7 GB | 68.8 GB | Runs comfortably | 95K | 2.01 |
| Q6_K | 6.56 | 53.9 GB | 57.0 GB | Runs comfortably | 128K | 2.44 |
| Q5_K_M | 5.67 | 46.6 GB | 49.7 GB | Runs comfortably | 128K | 2.81 |
| Q5_K_S | 5.52 | 45.3 GB | 48.4 GB | Runs comfortably | 128K | 2.88 |
| Q4_K_M | 4.85 | 40.0 GB | 43.1 GB | Runs comfortably | 128K | 3.26 |
| Q4_K_S | 4.58 | 37.8 GB | 40.9 GB | Runs comfortably | 128K | 3.43 |
| AWQ 4-bit | 4.25 | 37.8 GB | 40.9 GB | Runs comfortably | 128K | 3.44 |
| GPTQ 4-bit | 4.25 | 37.8 GB | 40.9 GB | Runs comfortably | 128K | 3.44 |
| MXFP4 | 4.25 | 37.8 GB | 40.9 GB | Runs comfortably | 128K | 3.44 |
| Q4_0 | 4.55 | 37.6 GB | 40.7 GB | Runs comfortably | 128K | 3.46 |
| IQ4_XS | 4.25 | 35.2 GB | 38.3 GB | Runs comfortably | 128K | 3.68 |
| Q3_K_M | 3.91 | 32.5 GB | 35.6 GB | Runs comfortably | 128K | 3.98 |
| IQ3_M | 3.70 | 30.8 GB | 33.9 GB | Runs comfortably | 128K | 4.18 |
| IQ3_XXS | 3.06 | 25.7 GB | 28.8 GB | Runs comfortably | 128K | 4.97 |
| Q2_K | 2.63 | 22.3 GB | 25.4 GB | Runs comfortably | 128K | 5.69 |
| IQ2_XXS | 2.06 | 17.8 GB | 20.9 GB | Runs comfortably | 128K | 7.05 |
| IQ1_M | 1.75 | 15.3 GB | 18.4 GB | Runs comfortably | 128K | 8.10 |