Tokens

Tokens

Image Generation

Our system uses Tokens to generate images. For a square (1:1) image, you’ll use 4 Squire 5 Tokens, while a wider (16:9) image requires 8 Squire 5 Tokens.

Text Generation

In our platform, text generation is facilitated through the use of tokens, which are units of computational effort required to process and generate text using the AI.

Understanding Tokens in Text Generation:

  • Input Tokens: These are the tokens that correspond to the text you input into the system. This includes the commands, prompts, or any data that you provide for processing.
  • Output Tokens: These tokens are consumed when the system generates a response to your input. The length and complexity of the generated text can affect the number of tokens used.

Calculating Usage for 1K Characters:

First, understand that 1,000 characters are approximately 0.5 Squire 5 Tokens because the AI counts tokens (these are distinct from Squire 5 Tokens), which can be parts of words, whole words, or punctuation, and not just characters. The AI also counts tokens for input (any instructions or text sent to it to process) and output (the part you get as a response). The exact token count can vary depending on the complexity of the text.

Therefore, for a request and a response that both have 1,000 characters each, the total cost would be the sum of both input and output costs, or 1 Squire 5 Token.

Please note that these calculations are approximations and the actual token count should be confirmed for precise costings, as the number of tokens can vary with the text's complexity.

As the cost of compute decreases, your Squire 5 Token value will increase. We will always pass on lower rates so that you can create more and build better characters and worlds.

Credits: Squire 5