llamaperf

AMD MI50 32GB vs CMP 170HX

For running local LLMs · 9 reports across 4 models

Side A
AMD MI50 32GB
Vendor
amd
VRAM
32GB
Memory
Discrete
Side B
CMP 170HX
Vendor
nvidia
VRAM
8GB
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 →

ModelAMD MI50 32GBCMP 170HX
Qwen3.6up to 35Bn=1210.0n=2
Qwen3.8up to 2400B15.0n=2n=1
DeepSeek V4 Flashup to 285B15.5n=129.0n=1
Kimi K2.69.7n=1

More comparisons