hjLabs AI Playground
Starting the on-device runtime…
DINOv2 Small is a 22.1M-parameter ONNX model that runs as an ONNX graph in the browser through transformers.js. On this page it runs entirely inside your browser: the 25 MB download comes straight from the Hugging Face repository “Xenova/dinov2-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.
Meta's self-supervised vision backbone. It outputs a 384-d vector per image — the features a classifier head is fitted on. In practice people reach for a model like this for experimenting with an open model without an API key or a server.
Available precisions: q8 (8-bit) and fp32 (full precision). q8 comes first because it roughly halves the download at a quality cost you will struggle to measure on this class of model; fp32 is there when you want the reference numbers. It is this playground's default pick for the task, and Hugging Face records about 17k downloads a month for the repository — a good sign that it is maintained and that the ONNX export is not abandoned.
Because there is no inference server behind this page, there is nothing to rate-limit and nothing to bill: the only cost of running DINOv2 Small here is the one-off 25 MB download, and once that is cached the page keeps working with the network switched off.
dinov2 smalldinov2 small onnxdinov2 small browserrun dinov2 small locallyxenova/dinov2-smallteach a classifier modeldinov2 small webgpudinov2 small download sizefree teach a classifier modeltransformers.js model
About 25 MB for the q8 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/dinov2-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.