Token Sayacı
Metninizin kaç token kullandığını, modelin bağlam penceresinin ne kadarını doldurduğunu ve göndermenin ne kadara mal olduğunu hesaplayın — tarayıcınızda, hiçbir şey yüklemeden.
Her şey tarayıcınızda hesaplanır. Metniniz hiçbir zaman yüklenmez, kaydedilmez ve saklanmaz.
0% of the context window
Yoğunluk, bulduğu yazı sistemine — ya da seçtiğiniz dile — göre uyarlanır.
Tahmin nasıl hesaplanıyor
- Paste or type your text into the box. The count updates as you type; there is no button to press and nothing to submit.
- Read the estimated token count. Characters, words and tokens are shown together so you can sanity-check the ratio for your language.
- Choose a model family. The context bar shows how much of that model’s window your text would fill.
- Check the indicative cost. Cost is shown per one thousand tokens at the rate you set, so you can plan a budget before you build anything.
Bu bir tahmindir, fatura tutarı değildir. Exact counts need each model’s own tokenizer vocabulary. This tool approximates from character classes, word shape and script — accurate enough to plan a budget or check whether a prompt fits, not accurate enough to reconcile an invoice.
Frequently asked questions
What is a token?
A token is the unit a language model reads and bills by. It is usually a word fragment rather than a whole word — common English words are often a single token, while longer or rarer words split into several.
How many characters is one token?
For ordinary English prose, roughly four characters per token. Code, punctuation-heavy text and non-Latin scripts use more tokens for the same number of characters, which is why this tool adjusts its estimate by script.
Why does Turkish use more tokens than English?
Most tokenizers were fitted mainly on English text, so Turkish words are split into more, smaller pieces. The same sentence can cost noticeably more tokens in Turkish than in English.
What is a context window?
The maximum number of tokens a model can consider at once, counting your input and its reply together. When a conversation exceeds it, the earliest parts stop being visible to the model.
Is this count exact or an estimate?
It is an estimate. Exact counts require each model’s own tokenizer vocabulary; this tool approximates from character classes and script, which is close enough for planning and budgeting but should not be used for billing reconciliation.
Is my text uploaded anywhere?
No. The calculation runs in your browser with JavaScript. Nothing you paste is sent to Leopati or to any third party, and nothing is stored.
Bir prompt'tan daha büyük bir şey mi planlıyorsunuz?
Leopati is an AI-native video production studio. If you are working out what an AI music video, film or commercial would take, the same honesty applies there: real numbers, before you commit.
Projeyi başlatın Yapay zekâ video prodüksiyonu ne kadar tutar Seçili işler