Live, local token counts and exact page selection inspection.
Users
4
store count
Rating
0.0 / 5
unchanged in 30 days
Ratings
0
total ratings
Store rank
#283K
#234K among extensions
Version
1.1.2
updated 22 Sep 2026
Users
monthly, 1 months
Latest, 16 Oct 2026:4 usersMove the mouse over the chart for any month
Ratings count
monthly, 1 months
Latest, 16 Oct 2026:0 ratingsMove the mouse over the chart for any month
The Chrome Web Store rounds user counts down to the leading digit, so the user line moves in steps. The ratings count is exact and is the better signal of real adoption.
About Statement Token Selector
See the small pieces behind every sentence.
Statement Token Selector turns selected text into a colorful, exact token breakdown—right where you read. Explore how words, spaces and punctuation become tokens, with live counts and clear details.
SELECT. EXPLORE. UNDERSTAND.
• Select text on supported webpages for live token and character counts.
• Click a sentence to inspect it, or drag to choose your own passage.
• See a floating token preview and pin colorful highlights when exact page mapping is available.
• Inspect token IDs and UTF-8 bytes, then copy text, IDs or JSON.
• Choose o200k_base, cl100k_base or r50k_base encoding.
BRING YOUR PDFs
Open the included PDF token reader, choose a local PDF, and explore selectable text with live previews, colored highlights, page navigation and zoom.
PRIVATE BY DESIGN
Processing stays on your device using bundled assets. No account, API key, analytics or upload is required. Selected text and PDFs stay in memory without extension storage. Explicit copy actions write the requested result to your clipboard.
GET STARTED
Click the toolbar icon on an ordinary webpage, then select text. Sentence mode starts enabled and can be turned off. Paste-to-analyze and system, light and dark appearance are also included.
A FEW LIMITS
Protected browser pages and inaccessible frames cannot be highlighted. Chrome’s built-in PDF viewer supports right-click “Analyze selected text” for panel results; inline PDF highlights require the included reader. Scanned PDFs without selectable text and password-protected PDFs are unsupported. Analysis is limited to 100,000 Unicode code points; token buttons and inline rendering are capped for large results. Counts describe selected text, not chat formatting or API billing overhead.
Uses community tiktoken JavaScript/WASM bindings with bundled encoding data, checked against official Python tiktoken fixtures. An independent extension, not an official OpenAI product.
New Version: 1.1.2 : fixes some of the bugs that were encountered during local testing.