llamaperf

Radeon AI PRO R9700 32GB vs RTX 3090

For running local LLMs · 92 reports across 11 models

Side A
Radeon AI PRO R9700 32GB
Vendor
amd
VRAM
32GB
Memory
Discrete
Side B
RTX 3090
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 →

ModelRadeon AI PRO R9700 32GBRTX 3090
Qwen3.8up to 2400B280.0n=12382.0n=45
Qwen3.6up to 35B84.5n=3104.0n=12
DeepSeek V4 Flashup to 285B60.0n=9
DeepSeek V4up to 284B51.1n=115.0n=1
GLM-5.2up to 744B7.3n=2
Qwen2.5up to 32B70.0n=2
DeepSeek V3n=1
GLM-4.5-Airup to 106Bn=1
GLM-5.355.0n=1
Museup to 30B94.0n=1
Qwen3up to 235B7.5n=1

More comparisons