llamaperf

H100 80GB vs RTX 4090

For running local LLMs · 6 reports across 3 models

Side A
H100 80GB
Vendor
nvidia
VRAM
80GB
Memory
Discrete
Side B
RTX 4090
Vendor
nvidia
VRAM
24GB
Memory
Discrete

Tokens per second by model

ModelH100 80GBRTX 4090
Qwen3.6up to 35B45.0n=180.0n=2
Gemma 4up to 31B125.3n=1149.6n=1
Qwen3.5up to 27Bn=1

More comparisons