Text Chunker
Split text into bounded character chunks with optional overlap.
Processing stays in your browser. Uses Unicode code points rather than model tokens. It preserves exact text and does not estimate a model's context limit.
How to use the text chunker
Split text into bounded character chunks with optional overlap. Complete the source text, chunk size (characters), overlap (characters) fields above, or upload a UTF-8 text file for the first field. Choose Run tool, review the output, then copy or download the result.
Example
A size of 10 with overlap 2 starts each successive chunk eight characters later.
Good to know
Uses Unicode code points rather than model tokens. It preserves exact text and does not estimate a model's context limit.