Arrange uses all-MiniLM-L6-v2, quantized to int8 and run locally.

Model: sentence-transformers/all-MiniLM-L6-v2 (Apache License 2.0)
https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2
ONNX conversion: Xenova/all-MiniLM-L6-v2
https://huggingface.co/Xenova/all-MiniLM-L6-v2/tree/751bff37182d3f1213fa05d7196b954e230abad9

Transformers.js: Apache License 2.0
https://github.com/huggingface/transformers.js
Full license: /transformers-license.txt

ONNX Runtime: MIT License
https://github.com/microsoft/onnxruntime
Full license: /onnx-license.txt

Notes, embeddings, query text, and positions stay on this device.