What actually runs well on a Raspberry Pi 5?
Our compute tiers are estimates based on parameter counts. Real numbers beat estimates.
If you have measured frame rates or latency on a Pi 5 for small detectors, depth models, compact VLMs or streaming speech, please share the model, runtime and settings.
Referenced repos
Indexed from Hugging Face and linked upstream, not hosted here.
Lightweight monocular model that predicts a relative depth map from a single RGB frame.
Small real-time DETR-style detector for the 80 COCO classes that refines box edges as distributions for tighter localization.
Half-billion-parameter vision-language model from Hugging Face that answers questions about images and short video clips.
The small size of the Moonshine Streaming family, an English-only encoder-decoder recognizer that uses a sliding-window encoder so it can transcribe live audio with low latency on edge hardware.