llamaperf

RTX 4080 Super vs RTX 4090

For running local LLMs · 18 reports across 4 models

Side A
RTX 4080 Super
Vendor
nvidia
VRAM
16GB
Memory
Discrete
Side B
RTX 4090
Vendor
nvidia
VRAM
24GB
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 4080 SuperRTX 4090
Qwen3.8up to 2400B59.0n=1134.0n=7
DeepSeek V4 Flashup to 285B180.0n=6
Gemma 4up to 31B149.6n=2
Qwen3.6up to 35B80.0n=2

More comparisons