What is a Token?
A Token is the smallest unit of text processed by GPT models. Roughly speaking, 1 English word ≈ 1.3 Tokens, 1 Chinese character ≈ 1.5-2 Tokens. Generally, 1000 Tokens ≈ 750 English words ≈ 500 Chinese characters.
What are the prices for GPT-4 and GPT-4o?
GPT-4: Input $0.03/1K Tokens, Output $0.06/1K Tokens. GPT-4o: Input $0.005/1K Tokens, Output $0.015/1K Tokens. GPT-3.5: Input $0.0005/1K Tokens. The above are reference prices, subject to OpenAI official pricing.
How accurate is the Token Calculator?
This tool uses an estimation algorithm based on BPE tokenization rules, with an accuracy of about 95% for common Chinese and English text. For exact Token counts, please refer to OpenAI's official tiktoken library. For text containing large amounts of code or special symbols, the estimation error may be slightly larger.
Why is my Token consumption higher than expected?
Common reasons: 1) Chinese consumes more Tokens than English; 2) Conversation history (context) accumulates; 3) System Prompt also takes up Tokens; 4) Special symbols and emojis consume more. It is recommended to regularly clear conversation context to save Tokens.