LLM token cost calculator
Turn your input and output token counts into a real dollar cost across the latest Claude, GPT and Gemini models.
Cost per call
$0.600
Estimated monthly cost
$600.00
1,000 calls · Claude Fable 5.1
- Input cost / call$0.500
- Output cost / call$0.100
Standard (non-batch) list rates per million tokens, USD, June 2026. See all pricing data
Prices and models updated automatically on October 18, 2026 (LiteLLM).
Frequently asked questions
- How is the cost calculated?
- Cost = (fresh input tokens ÷ 1M × input price) + (cached tokens ÷ 1M × cache-read price) + (output tokens ÷ 1M × output price), times your calls per month.
- What is cached input?
- Providers bill already-cached input tokens at roughly 10% of the base price. A stable context (system prompt, an unchanged CLAUDE.md across turns) earns that discount. See our prompt caching page.
- Are the prices current?
- Yes: the model list and their standard (non-batch) public list rates are fetched automatically every day from LiteLLM's open price table, so a new model shows up as soon as it is listed there. Still check with the provider before quoting a production cost model.
- How do I reduce this cost?
- Most of an AI agent's cost comes from repeated input tokens (file re-reads, tool output). Trimming those before they reach the model — what Tokenade does — directly lowers the bill.
Stop counting. Let Tokenade cut the bill.
Tokenade automatically trims your AI coding agent's tokens. Open source (MIT/Apache-2.0), 100% local, zero telemetry.