M4 Pro 48GB vs RTX 5090
For running local LLMs · 70 reports across 12 models
Highest reported generation speed by model family
These maxima come from different community setups. Model size, quantization, GPU count, offloading and concurrency may differ. They do not establish which GPU is faster for the same workload. Check the comparison conditions →
| Model | M4 Pro 48GB | RTX 5090 |
|---|---|---|
| Qwen3.8up to 2400B | 20.3n=1 | 220.0n=35 |
| DeepSeek V4 Flashup to 285B | — | 120.0n=10 |
| Museup to 30B | 18.0n=2 | 253.0n=4 |
| Qwen3.6up to 35B | — | 215.1n=6 |
| Gemma 4up to 31B | — | 578.0n=4 |
| GLM-5.3up to 321B | 1.8n=1 | 24.5n=1 |
| GLM-5.2up to 744B | — | —n=1 |
| Kimi K2.5 | — | —n=1 |
| LingBot-World-V2up to 1B | —n=1 | — |
| Qwen2.5up to 32B | — | —n=1 |
| Qwen3up to 235B | — | 159.0n=1 |
| Qwen3.5up to 122B | — | —n=1 |