LLM API Cost Calculator and Comparison —
Claude vs ChatGPT vs DeepSeek

API costs are calculated based on the number of tokens processed. You are charged separately for Input tokens (the prompt you send) and Output tokens (the response the AI generates). Prices vary significantly between models.

A token is a piece of a word. As a general rule of thumb, 1 token ≈ 4 characters in English, or about ¾ of a word. Therefore, 100 tokens are roughly 75 words.

Generating text (output) is significantly more computationally intensive for the AI provider than simply reading text (input). This is why output tokens generally cost 3x to 6x more than input tokens.

This calculator uses the official, published pricing from OpenAI, Anthropic, Google, and DeepSeek as of May 2026. However, your actual costs may vary slightly depending on how each provider counts tokens in practice, especially with image inputs or tool calling.

A context window is the maximum number of tokens a model can process in a single request. This includes both your input prompt and the model's generated output. Models with larger context windows (like Gemini 3 Pro with 2M tokens) can analyze entire books or long videos at once.

Currently, DeepSeek V4 and Google Gemini 2.5 Flash offer some of the most competitive pricing for high-volume data processing and basic tasks, costing pennies per million tokens compared to frontier models.

Yes! Many providers like Anthropic and OpenAI now offer prompt caching. If you repeatedly send the same large context (like a codebase or long document), the cached input tokens are usually discounted by 50% to 80%.

A typical customer service chatbot interaction might use around 500-1,000 input tokens (including the system prompt and conversation history) and generate 100-300 output tokens per message.