| Machine | Fits? | Tokens/sec | |
|---|---|---|---|
| Mac Studio M5 Ultra (96GB) | Yes | 117 | Check price → |
| Mac Studio M5 Ultra (256GB) | Yes | 117 | Check price → |
| Mac Studio M5 Max (40-core GPU, 64GB) | Yes | 59 | Check price → |
| Mac mini M5 Pro (24GB) | Yes | 29 | Check price → |
| Mac mini M5 Pro (64GB) | Yes | 29 | Check price → |
| Mac mini M4 Pro (64GB) older | Yes | 26 | Check price → |
| NVIDIA DGX Spark (128GB) | Yes | 23 | Check price → |
| Ryzen AI Max+ 395 mini PC (128GB) | Yes | 21 | Check price → |
| Mac mini M6 (16GB) | Yes | 16 | Check price → |
| Mac mini M6 (32GB) | Yes | 16 | Check price → |
| Mac mini M4 (16GB) older | Yes | 11 | Check price → |
| Mac mini M2 (16GB) older | Yes | 9.5 | Check price → |
| Beelink SER7 (Ryzen 7 7840HS, 32GB) | Yes | 6.4 | Check price → |
| Minisforum UM790 Pro (Ryzen 9 7940HS, 32GB) | Yes | 6.4 | Check price → |
| GEEKOM A6 (Ryzen 7 6800H, 32GB) | Yes | 5.5 | Check price → |
Around 10 tokens/sec reads comfortably; under 5 feels slow. How we estimate.
Other olmo2 sizes
Embed this table
Free for blogs, READMEs and model cards. Updates automatically.
<iframe src="https://vsmacs.com/embed/olmo2-13b" width="100%" height="420" style="border:0" loading="lazy" title="OLMo 2 13B speed by machine"></iframe>