Contador de tokens

Cuenta tokens para modelos LLM.

~95 Tokens

315 Chars

Prompt & Context Input

Token & Cost Breakdown

Estimated Tokens

~3.32 chars/token

Word Count

Total words

Context Window Usage0.07% of 128k

Input Prompt Cost

$0.000238

Output Gen Cost

$0.00095

Cross-Model Pricing & Token Capacity Comparison

Model Name	Provider	Context Window	Estimated Tokens	Input Cost ($)	Output Cost ($)
OpenAI GPT-4o	OpenAI	128k tokens	95	$0.000238	$0.00095
Anthropic Claude 3.5 Sonnet	Anthropic	200k tokens	99	$0.000297	$0.001485
Google Gemini 1.5 Pro	Google	2000k tokens	90	$0.000113	$0.00045
DeepSeek V3	DeepSeek	128k tokens	97	$0.000014	$0.000027
Meta Llama 3.3 70B	Meta / Open Source	128k tokens	95	$0.000033	$0.000038

Understanding BPE Tokenization in Modern LLMs

Large Language Models (LLMs) do not read text character-by-character. Instead, they process text in chunks called tokens using Byte-Pair Encoding (BPE) or SentencePiece algorithms.

English Words: 1 token is approximately 4 characters or 0.75 words.
Code & JSON: Special symbols ({}[]:;=&) frequently count as individual tokens.
Multi-lingual Text: Accented or non-Latin alphabets can consume multiple byte tokens per character.