hjLabs AI Playground
Starting the on-device runtime…
OWL-ViT Base 32 is a 154M-parameter ONNX model that locates objects described by free-text labels you supply at query time, with no retraining. On this page it runs entirely inside your browser: the 159 MB download comes straight from the Hugging Face repository “Xenova/owlvit-base-patch32” into your browser cache, and every inference after that executes on your own GPU through WebGPU. That is a one-off download of well under a minute on a normal connection, and small enough that integrated graphics handle it without complaint.
Type any phrase and it draws a box — the fastest and most forgiving of the open-vocabulary detectors. In practice people reach for a model like this for finding things no fixed class list covers — a specific product, a logo, an unusual piece of equipment.
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 3k 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 OWL-ViT Base 32 here is the one-off 159 MB download, and once that is cached the page keeps working with the network switched off.
owl-vit base 32owl-vit base 32 onnxowl-vit base 32 browserrun owl-vit base 32 locallyxenova/owlvit-base-patch32find anything modelowl-vit base 32 webgpuowl-vit base 32 download sizefree find anything modeltransformers.js model
About 159 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/owlvit-base-patch32 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.