Token counter & cost estimator
Paste any text to estimate how many tokens it is, and what that costs as input or output across the major models. The count is an estimate: exact numbers depend on each model's tokenizer, but this gets you within a few percent for planning.
| Model | As input | As output | Monthly (in) |
|---|
Approximate public figures as of mid-2026; prices change often. Verify live provider pricing before relying on these numbers.
Why token counts matter
Everything you send and receive is billed per token, and the parts people forget are the system prompt and context that ride along on every call. If a model is talkative, its output tokens can quietly dominate the bill even when the per-token price looks low, which is exactly the trap covered in the token economy. To turn a per-call number into a per-finished-task number, including retries and loops, use the agent cost simulator or the LLM cost calculator.