API Cost Calculator

🟡Intermedio

✅

Última verificación: 2026-06-21.

Enter the current per-million-token prices yourself from the official pricing page — we don't hard-code them because they change.

Confirma con la fuente oficial.

Estimate the monthly cost of an API workload, and see how much a stable cached prefix saves. All client-side — nothing leaves your browser.

Input tokens / request

Output tokens / request

Requests / month

Input price

$ / 1M tok

Output price

$ / 1M tok

Cached input

% of prefix

$105.00estimated / month

Input (fresh)	$45.00
Output	$60.00

Enter the current per-million-token prices from the official pricing page (we don't hard-code them — they change). The cached-read rate is model-dependent (~0.1× input here is illustrative). Estimate only — verify with Anthropic's token counting and pricing.

:::tip Cut the bill, not the quality The biggest levers: right-size the model, trim input tokens, and reuse a stable prefix with prompt caching. Full strategy in Cost & Latency. :::

Next

• Tokens, Context & Pricing — how cost is calculated
• Prompt Caching & Cost Optimization
• Current Models & Pricing