hjLabs AI Playground
Starting the on-device runtime…
MobileViT Small is a 5.6M-parameter ONNX model that assigns an image to one or more labels from the classes it was trained on. On this page it runs entirely inside your browser: the 23 MB download comes straight from the Hugging Face repository “Xenova/mobilevit-small” into your browser cache, and every inference after that executes on your own GPU through WebGPU. At this size the download is over before you have finished reading this paragraph, even on a phone tethered to a hotspot, and it costs almost nothing to keep cached alongside a dozen other models.
Smallest classifier here at 23 MB, designed for phones — the right default on mobile data. In practice people reach for a model like this for sorting and filtering image libraries, moderation gates, and quick quality checks on user uploads.
Available precisions: fp32 (full precision) and q8 (8-bit). fp32 comes first because at this size full precision is both the fastest WebGPU path and the best quality; the quantized variant exists for the WebAssembly fallback and for tight cache budgets. Hugging Face records about 976 downloads a month. It is a specialist choice rather than a default, which is exactly why it is worth trying when the popular model gets your particular input wrong.
Because there is no inference server behind this page, there is nothing to rate-limit and nothing to bill: the only cost of running MobileViT Small here is the one-off 23 MB download, and once that is cached the page keeps working with the network switched off.
mobilevit smallmobilevit small onnxmobilevit small browserrun mobilevit small locallyxenova/mobilevit-smallimage classification modelmobilevit small webgpumobilevit small download sizefree image classification modeltransformers.js model
About 23 MB for the fp32 build, including the tokenizer and config files the runtime also has to fetch — this is the whole payload, not just the weights file. It is downloaded once and then served from the browser Cache Storage API.
Yes, through the WebAssembly fallback, which is slower but works anywhere. WebGPU (Chrome or Edge 113+, Safari 26+) is what makes it feel interactive. On WASM, prefer the smaller quantized build.
No. Xenova/mobilevit-small is a public repository, so your browser fetches the ONNX files directly over HTTPS. There is no account, no key and no server in between.
No. The model weights are downloaded from Hugging Face to your browser once, and every inference after that runs on your own GPU through WebGPU. Your text, images and audio are never sent anywhere — hjLabs.in has no inference server and no way to see your input.