The paper states: 'the same SmolLM2 model family reaches 8.42 tokens/s on Orange Pi 5 Pro but 64.38 tokens/s on an RTX 3090-class GPU.'
正在加载行业信息...
正在加载行业信息...
开源项目
更新于 2026/7/28 17:28:01
A model family evaluated in the study for cross-device latency profiling.
The paper states: 'the same SmolLM2 model family reaches 8.42 tokens/s on Orange Pi 5 Pro but 64.38 tokens/s on an RTX 3090-class GPU.'
暂无来源链接。