hjLabs AI Playground
Starting the on-device runtime…
MobileBERT SQuAD2 is a 25M-parameter ONNX model that finds and returns the exact span of a passage that answers a question. On this page it runs entirely inside your browser: the 28 MB download comes straight from the Hugging Face repository “onnx-community/mobilebert-uncased-squad-v2-ONNX” 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. It is an English-language model; expect quality to fall off sharply on other languages.
The smallest QA model here at 28 MB — good enough for FAQ lookup on a phone. In practice people reach for a model like this for pulling a specific fact out of a contract, a manual or a long document with a citable span rather than a paraphrase.
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. Hugging Face records about 183 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 MobileBERT SQuAD2 here is the one-off 28 MB download, and once that is cached the page keeps working with the network switched off.
mobilebert squad2mobilebert squad2 onnxmobilebert squad2 browserrun mobilebert squad2 locallyonnx-community/mobilebert-uncased-squad-v2-onnxquestion answering modelmobilebert squad2 webgpumobilebert squad2 download sizefree question answering modeltransformers.js model
About 28 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. onnx-community/mobilebert-uncased-squad-v2-ONNX 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.