
























Token Studio
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
Loading tokenizer…
6.4 MB · one-time
327characters
270no spaces
58words
4sentences
Token breakdown
GPT-5 · o200k_base
Loading GPT tokenizer · 6.4 MB
Compare models — same text 3 of 25 · click to inspect
Model Tokens Cost
Claude Opus
loading tokenizer…
—
Gemini Pro
loading tokenizer…
—
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.
此内容由惯性聚合(RSS阅读器)自动聚合整理,仅供阅读参考。 原文来自 — 版权归原作者所有。