llamaperf

H100 80GB vs RTX 3080 20GB

For running local LLMs · 9 reports across 5 models

Side A
H100 80GB
Vendor
nvidia
VRAM
80GB
Memory
Discrete
Side B
RTX 3080 20GB
Vendor
nvidia
VRAM
20GB
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 →

ModelH100 80GBRTX 3080 20GB
DeepSeek V4 Flashup to 285B193.0n=117.5n=2
Qwen3.8up to 2400B57.5n=3
Gemma 4up to 31B125.3n=1
Qwen3.5up to 122Bn=1
Qwen3.6up to 35B45.0n=1

More comparisons