How LLM pricing works
Providers bill per million tokens, with separate input (your prompt and context) and output (the model's response) prices — output tokens usually cost several times more. Monthly cost = (input tokens × input price + output tokens × output price) ÷ 1,000,000.
Estimating your token counts
A token is roughly ¾ of an English word, so a 1,000-word prompt is ~1,300 tokens. Long system prompts and retrieved context are billed on every request, which is why input tokens dominate many workloads — trimming the prompt is often the biggest cost lever.
Frequently asked questions
What is a token?
Roughly three-quarters of a word in English text; 1,000 words is about 1,300 tokens.
Why are output tokens more expensive?
Generating text is more computationally expensive than reading it, so providers price output tokens several times higher than input.
Why enter prices manually instead of picking a model?
Model prices change frequently. Entering the current numbers from your provider's pricing page keeps the estimate accurate for any model, from any vendor.