hjLabs AI Playground
Starting the on-device runtime…
BART Large MNLI is a 407M-parameter ONNX model that scores text against categories you invent at query time by reframing the problem as natural-language inference. On this page it runs entirely inside your browser: the 415 MB download comes straight from the Hugging Face repository “Xenova/bart-large-mnli” into your browser cache, and every inference after that executes on your own GPU through WebGPU. Expect a download of roughly a minute on a decent connection. It fits comfortably in the browser cache and in the memory of any GPU from the last few years, including laptop integrated graphics. It is an English-language model; expect quality to fall off sharply on other languages.
The model that popularised zero-shot classification. Most accurate here and, at 415 MB, the slowest. In practice people reach for a model like this for intent routing, topic labelling and triage where there is no labelled training set and never will be.
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 around 67k downloads a month — steady, real usage rather than a one-off research drop.
Because there is no inference server behind this page, there is nothing to rate-limit and nothing to bill: the only cost of running BART Large MNLI here is the one-off 415 MB download, and once that is cached the page keeps working with the network switched off.
bart large mnlibart large mnli onnxbart large mnli browserrun bart large mnli locallyxenova/bart-large-mnlizero-shot text labels modelbart large mnli webgpubart large mnli download sizefree zero-shot text labels modeltransformers.js model
About 415 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/bart-large-mnli 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.