Needle-rs – AI Function calling in the browser, 258 KB WASM

geekgineer1 pts0 comments

needle-rs — tool-calling AI, 258 KB, in your browser

model not loaded

Load model (22 MB, cached after first run)

Downloading weights…<br>0 / 22 MB

Weights are streamed from<br>HuggingFace<br>and cached by your browser. One-time download.

Try an example

Query

Tools (JSON)<br>OpenAI-compatible

[{<br>"name": "get_weather",<br>"description": "Get current weather for a city",<br>"parameters": {<br>"type": "object",<br>"properties": {<br>"location": {"type": "string", "description": "City name"},<br>"unit": {"type": "string", "description": "celsius or fahrenheit"}<br>}]

Run inference →

Output

Use it in your own project

npm · needle-rs

crates.io · needle-infer

PyPI · needle-rs

GitHub · needle-rs

$ npm install needle-rs

npmjs.com ↗

$ cargo add needle-infer

crates.io ↗

$ pip install needle-rs

pypi.org ↗

needle browser description type calling model

Related Articles