OpenBMB says MiniCPM5-2B runs at 37 tokens/s on iPhone Air
Overview
OpenBMB reports that its MiniCPM5-2B model runs at 37 tokens per second on an iPhone Air, which it presents as evidence that small open multimodal models can run on phones without a cloud GPU. NobodyWho, which says the model is available in its Chat app, also highlighted the result.
The claim comes from OpenBMB's own post and the NobodyWho note; no independent benchmark of the speed was reported.
Written by AI from the articles below · updated Oct 9, 4:56 AM ET
Check the sources:
Article timeline
Follow the coverage from different perspectives. Times are ET.
OpenBMB@OpenBMBMiniCPM5-2B runs at 37 tok/s on iPhone AirAIOpenBMB reports that its MiniCPM5-2B model runs at 37 tokens per second on an iPhone Air. The post presents this as evidence that small open multimodal models can run on mobile devices without a cloud GPU, with NobodyWho noting the model is available in its Chat app.
Heat trend
Not enough continuous observations to show a trend yet.