Show HN: In-browser LLM token counter and cost estimation

angelmm1 pts0 comments

Token Counter and Cost | HolaClaw

Search Ctrl K Download 👋¡Hola!

Token Studio<br>Token Counter and Cost Estimation<br>Paste any text to see how it tokenizes across model families — with live token counts, cost estimates, and context-window usage. Everything runs locally in your browser.

Runs entirely in your browser Nothing is uploaded

Claude Opus Anthropic<br>Loading tokenizer…10.9 MB · one-time

Est. cost

Tokens / word

Chars / token

Characters<br>327<br>58 words

Context window

Input text<br>Sample Clear

HolaClaw runs your own AI assistant right on your Mac — give it a name, a personality, and a purpose. Luna remembers that you hate early flights and love quiet coastal towns, then plans trips that actually feel like you. No cloud accounts, no data leaving your machine. Just an assistant that thinks with you, every single day. 327characters<br>270no spaces<br>58words<br>4sentences

Token breakdown<br>Claude Opus · claude-v4

Loading Claude tokenizer · 10.9 MB

Compare models — same text 3 of 25 · click to inspect<br>Model Tokens Cost<br>Claude Opus<br>loading tokenizer…

GPT-5<br>loading tokenizer…

Gemini Pro<br>loading tokenizer…

Add a model to compare

Counts come from each model's real tokenizer, running entirely in your browser. Cost and context-window figures are estimates for guidance and may vary by model version.

token cost tokenizer model claude loading

Related Articles