llamaperf

H100 80GB vs Radeon AI PRO R9700 32GB

For running local LLMs · 20 reports across 6 models

Side A
H100 80GB
Vendor
nvidia
VRAM
80GB
Memory
Discrete
Side B
Radeon AI PRO R9700 32GB
Vendor
amd
VRAM
32GB
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 80GBRadeon AI PRO R9700 32GB
Qwen3.8up to 2400B280.0n=12
Qwen3.6up to 35B45.0n=184.5n=3
DeepSeek V4up to 284B51.1n=1
DeepSeek V4 Flashup to 285B193.0n=1
Gemma 4up to 31B125.3n=1
Qwen3.5up to 122Bn=1

More comparisons