hjLabs AI Playground
Starting the on-device runtime…
Kokoro 82M v1.1 (Chinese) is an 82M-parameter ONNX model that synthesises speech audio from written text. On this page it runs entirely inside your browser: the 127 MB download comes straight from the Hugging Face repository “onnx-community/Kokoro-82M-v1.1-zh-ONNX” 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. Declared languages: zh, en.
Kokoro retrained with a large Mandarin voice pack; also still speaks English. Runs through kokoro-js. In practice people reach for a model like this for voicing drafts, generating narration, and accessibility read-back — with no per-character billing.
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 1k 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 Kokoro 82M v1.1 (Chinese) here is the one-off 127 MB download, and once that is cached the page keeps working with the network switched off.
kokoro 82m v1.1 (chinese)kokoro 82m v1.1 (chinese) onnxkokoro 82m v1.1 (chinese) browserrun kokoro 82m v1.1 (chinese) locallyonnx-community/kokoro-82m-v1.1-zh-onnxtext to speech modelkokoro 82m v1.1 (chinese) webgpukokoro 82m v1.1 (chinese) download sizefree text to speech modeltransformers.js model
About 127 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/Kokoro-82M-v1.1-zh-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.