tesseract.js 7.0.0 + tesseract.js-core 7.0.0 — the Tesseract OCR engine compiled to WebAssembly, run in a Web Worker.
License: Apache License 2.0 (see LICENSE; tesseract.js-core's own LICENSE is the same Apache-2.0 text). The libraries compiled
into the core follow in LICENSE with their own texts (Leptonica, zlib, libpng, libjpeg/IJG, libtiff, libwebp; approved by the
owner 2026-10-26 for the OCR engine); NOTICE lists them and the JavaScript bundles' parts.
Source: https://github.com/naptha/tesseract.js  Package: https://www.npmjs.com/package/tesseract.js/v/7.0.0
        https://github.com/naptha/tesseract.js-core  Package: https://www.npmjs.com/package/tesseract.js-core/v/7.0.0
Kept, unchanged: dist/tesseract.esm.min.js, dist/worker.min.js (+ their .LICENSE.txt), and from the core only the three
LSTM-only builds as core/tesseract-core{,-simd,-relaxedsimd}-lstm.wasm.js (the worker picks one by the browser's SIMD support;
the legacy-engine builds are left out: MarkupCraft runs OEM LSTM_ONLY). Source maps left out.
Used by MarkupCraft (0.7.1 71.10) to read the words on scanned pages, only when someone runs Read text (OCR); worker, core
and language data are always given as local paths (workerPath, corePath, langPath), so nothing is fetched from a CDN.
The language data is a separate folder: assets/vendor/tessdata/.
