AI Token Cost Calculator

Price a Claude API workload before you run it — per call, per day, per month — with prompt-cache reads and Batch API discounts applied, across every current model.

Rates last verified 2026-07-17 against Anthropic's published pricing.

Estimated cost

per call
per day
per 30 days
Same workload on…$/MTok in / outper 30 days

Note: Claude Fable 5, Opus 4.7/4.8, and Sonnet 5 use a newer tokenizer that produces roughly 30% more tokens for the same text than Sonnet 4.6-era models — the comparison table applies that adjustment when you switch tokenizer families, and marks adjusted rows with †.

How Claude API pricing actually works

Every request bills two meters: input tokens (everything you send — system prompt, conversation history, tool definitions, documents) and output tokens (everything the model writes back, including its thinking). Output is 5× the input price on every current Claude model, so workloads that generate long answers from short prompts cost far more than the raw token counts suggest — and vice versa: stuffing a large document into a prompt for a one-line answer is cheaper than it looks.

Cache reads are the biggest lever

A prompt-cache read bills at 10% of the input price. A chat app that resends a 20,000-token system prompt on every turn pays full price once and a tenth of it on every subsequent turn — if the prefix is byte-stable. The calculator's cache slider models this: drag it to the share of your input that repeats across calls. Cache writes cost 1.25× (5-minute TTL) or 2× (1-hour TTL), so caching pays for itself after one to two reads.

Batch when nobody is waiting

The Batch API halves both meters in exchange for asynchronous delivery (most batches finish within an hour). Classification runs, nightly content generation, backfills — anything without a person watching — should be batched by default.

Context windows are a budget, not a feature

A 1M-token context window bills at the same per-token rate whether you use 9k or 900k of it. The window doesn't cost extra — filling it does. Long agentic sessions that resend their full history each turn grow quadratically in input spend unless caching or compaction is in play.

Paid tier — a cost-optimization report for your workload ($0.99)

Describe your application and monthly volume; get back a structured report: projected spend on the right model mix, what caching and batching would save you specifically, and the break-even points — priced from the same verified rate table this page uses.

What a real Claude app spends (with receipts)

Rate tables tell you the price of a token, not the cost of a system. We publish the companion guide What a Production Claude App Actually Costs to Run from our own production cost logs — every API call metered across a three-plane publishing pipeline — including the part no rate table shows: the model is usually the cheapest line item.