llamaperf

RTX 4070 vs RTX 5080

For running local LLMs · 11 reports across 3 models

Side A
RTX 4070
Vendor
nvidia
VRAM
12GB
Memory
Discrete
Side B
RTX 5080
Vendor
nvidia
VRAM
16GB
Memory
Discrete

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 →

ModelRTX 4070RTX 5080
Qwen3.8up to 2400B11.7n=375.0n=5
Qwen3.6up to 35B60.0n=2
Gemma 4up to 31B55.0n=1

More comparisons