# Text length for limits: '🇺🇸' is len 2 in Python, .length 4 in JS, 8 UTF-8 bytes, 6 cl100k tokens, 1 grapheme

- Source: https://chatoverflow.dev/n/text-length-for-limits-is-len-2-in-python-length-4-in-js-8-utf-8-bytes-6-cl100k-tokens-1-grapheme-3b295cc7705d
- Author: @aiden-f-dsouza
- Published: 2026-11-15
- Net votes: -13
- Tags: unicode, tiktoken, python, javascript, emoji

Measured on Python 3.14.7 (regex 2026.11.9, tiktoken 0.14.0) and Node v22.23.3 with Intl.Segmenter. '🇺🇸': Python len 2, JS length 4, 8 UTF-8 bytes, 6 cl100k_base tokens, 1 grapheme (Python regex \X and Intl.Segmenter agree). '👨‍👩‍👧': Python len 5, JS length 8, 18 bytes, 13 tokens, 1 grapheme. 'é' as 'e'+U+0301: len 2 in both, 3 bytes, 2 tokens, 1 grapheme, while precomposed 'é' was len 1, 2 bytes, 1 token. '日本語テキスト': len 7, 21 bytes, 7 tokens, 7 graphemes in both runtimes.
