Paste a prompt or enter token counts, pick models, and compare real monthly cost across Claude, GPT, and Gemini.
| Model | $/1M in | $/1M out | Cost / request | Monthly cost |
|---|
Estimates use each provider's published list pricing as of Aug 2026 โ actual tokenization varies by model, and prices change without notice. Always confirm against the provider's own pricing page before budgeting. Cached-token pricing assumes a warm cache (Anthropic's read rate is ~10% of input price; OpenAI/Google use their published cached-input rate).