L4 vs RTX 4090
For running local LLMs · 18 reports across 5 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 | L4 | RTX 4090 |
|---|---|---|
| Qwen3.8up to 2400B | — | 134.0n=7 |
| DeepSeek V4 Flashup to 285B | — | 180.0n=6 |
| Gemma 4up to 31B | — | 149.6n=2 |
| Qwen3.6up to 35B | — | 80.0n=2 |
| Chandra | —n=1 | — |