M2 Max 96GB vs RTX 3090
For running local LLMs · 16 reports across 3 models
Tokens per second by model
| Model | M2 Max 96GB | RTX 3090 |
|---|---|---|
| Qwen3.6up to 35B | 28.0n=3 | 104.0n=10 |
| Qwen2.5 | — | 70.0n=2 |
| DeepSeek V3 | — | —n=1 |
For running local LLMs · 16 reports across 3 models
| Model | M2 Max 96GB | RTX 3090 |
|---|---|---|
| Qwen3.6up to 35B | 28.0n=3 | 104.0n=10 |
| Qwen2.5 | — | 70.0n=2 |
| DeepSeek V3 | — | —n=1 |