M2 Ultra 192GB vs M5 Max 128GB
For running local LLMs · 7 reports across 4 models
Tokens per second by model
| Model | M2 Ultra 192GB | M5 Max 128GB |
|---|---|---|
| Gemma 4up to 31B | — | 7.5n=3 |
| Qwen3.6up to 35B | — | 5.5n=2 |
| DeepSeek V3 | 28.0n=1 | — |
| Qwen3.5up to 122B | — | —n=1 |
For running local LLMs · 7 reports across 4 models
| Model | M2 Ultra 192GB | M5 Max 128GB |
|---|---|---|
| Gemma 4up to 31B | — | 7.5n=3 |
| Qwen3.6up to 35B | — | 5.5n=2 |
| DeepSeek V3 | 28.0n=1 | — |
| Qwen3.5up to 122B | — | —n=1 |