Open-source licenses
Token Saver is built with these open-source libraries. Thank you to their authors.
| Library | Used for | License |
|---|---|---|
| pdf.js 3.11.174 (Mozilla) | Reading PDFs | Apache-2.0 |
| js-tiktoken 1.0.21 | Counting tokens (o200k_base) | MIT |
| Turndown 7.2.0 | HTML to Markdown | MIT |
| turndown-plugin-gfm 1.0.2 | Tables in Markdown | MIT |
| Readability 0.6.0 (Mozilla) | Finding the main content of a page | Apache-2.0 |
| mammoth 1.13.0 | Reading Word documents | BSD-2-Clause |
| read-excel-file 9.3.10 | Reading Excel spreadsheets | MIT |
| tesseract.js 7.0.0 and tesseract.js-core 7.0.0 | Reading text in screenshots (OCR) | Apache-2.0 |
| Tesseract trained data (English, Spanish) | OCR language models | Apache-2.0 |
The libraries are included unmodified, apart from bundling and minification. The full license texts are linked above.
This website
The code of Token Saver itself is released under the MIT License.