Estimate Anthropic API charges by pricing uncached input, cache writes, cache reads, and output separately, then account for any feature charges or pricing modifiers. After calculating usage, check whether it will be paid from Console credits or billed under a sales-arranged invoice. Credits affect how a balance is paid; they do not change the token rates used to calculate usage.
Calculate each token category at its own rate
A single total-token figure will not produce a reliable estimate when prompt caching is involved. Use the selected model’s current input and output rates, and count uncached input, cache creation, cache reads, and generated output separately. Anthropic’s pricing page lists current rates and cache pricing; rates and model-specific exceptions can change.
For a model using the standard cache multipliers, the request estimate is:
cost = (uncached_input_tokens × I + 5m_cache_write_tokens × 1.25I + 1h_cache_write_tokens × 2I + cache_read_tokens × 0.1I + output_tokens × O) / 1,000,000
Free tools Windows power users keep installed
One-click scans. No signup required.
#1 Best Overall
Here, I is the model’s base input price per million tokens and O is its output price per million tokens. Use the live pricing page’s multiplier if the selected model has a cache-read exception rather than assuming all models use 0.1×.
- Uncached input: multiply tokens not served from cache by the model’s input rate.
- Cache creation: price tokens written to the 5-minute and 1-hour caches separately. Anthropic documents standard rates of 1.25× and 2× base input, respectively.
- Cache reads: apply the model’s listed read rate. The standard rate is 0.1× base input.
- Output: multiply generated output tokens by the model’s output rate.
- Other charges: add separately metered features and account for any applicable pricing modifiers.
This is a request-cost template, not an invoice calculator. It estimates the listed token and feature charges; your account’s billing arrangement, adjustments, and usage controls must be checked separately.
Rank #2
Decide whether cache reuse is likely to pay off
Caching adds a write premium but can reduce the cost of later reads. At Anthropic’s standard multipliers, the savings on cache reads offset the 5-minute write premium after one cache read, and the 1-hour write premium after two. This is a break-even rule based on the documented multipliers and otherwise comparable token pricing, not a guarantee of savings for every workload.
Before relying on that rule, estimate how many times the same content will actually be reused while its cache remains available. The model’s cache-read rate, cache duration, and pattern of reuse determine whether the initial write cost is recovered.
Rank #3
Turn a request estimate into a monthly forecast
- Choose representative requests for the workload and estimate uncached input, each cache-write duration, cache reads, and output for each one.
- Apply the model’s current rates and any feature charges or modifiers to each request estimate.
- Multiply the per-request category totals by expected monthly request volume. Allow for variation in prompt size, output length, cache reuse, and feature usage.
- Keep the estimated usage charge separate from credits or invoice payments. Do not assume a recurring credit amount from a plan name or a general pricing page.
For an illustration, Anthropic’s pricing documentation gives a Managed Agents example in which 50,000 input tokens at $5 per million, 15,000 output tokens at $25 per million, and one hour of Managed Agents runtime at $0.08 total $0.705. Counting 40,000 input tokens as cache reads at the stated rate changes the total to $0.525. This example includes Managed Agents runtime; it is not a bare Messages API request price.
Understand what “monthly credits” means for your account
Anthropic describes different payment arrangements that should not be treated as the same kind of monthly allowance. Console API usage is generally funded by prepaid credits, while organizations with a sales-arranged invoicing agreement are billed in arrears. Optional usage credits for certain paid Claude plans are a separate arrangement again.
| Arrangement | How it works | What to verify |
|---|---|---|
| Console API and playground prepaid credits | Most organizations buy credits before using the API or playground; usage is charged against those credits at current pricing. Purchased credits are available immediately and expire one year after purchase. If they run out, API calls and playground use stop until more credits are added. | Check the Console Billing page for available balance, usage, expiration, and spending controls. |
| Sales-arranged monthly invoicing | Eligible organizations with an arrangement established through Anthropic Sales are billed monthly in arrears for aggregate usage at standard pay-as-you-go prices. This is an invoicing arrangement, not a recurring credit grant. | Check whether invoicing is configured for your organization and review its invoices. |
| Paid Claude plan usage credits | Anthropic’s Help Center describes optional usage credits for certain paid Claude plans after included usage limits. These credits are distinct from included plan limits and are billed at standard API rates; availability and expiration can depend on plan terms and jurisdiction. | Check the relevant Claude account’s billing terms and balance. Do not assume these credits are the same as Console prepaid credits. |
Anthropic’s documentation does not establish a universal monthly credit amount or your personal balance, eligibility, or next expiration date. Those details depend on the account and its arrangement. See the Anthropic Help Center guidance on API access and credits and confirm the relevant account settings.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Reconcile estimates against actual usage
Once usage occurs, compare the estimate’s token categories with Anthropic’s usage data rather than looking only at an overall token total. The usage-report endpoint documentation describes breakdowns for uncached input, 5-minute cache creation, 1-hour cache creation, and cache reads. Compare those reported amounts with the forecast, then check the Console Billing page for available credits and usage or review the invoice if the organization is billed monthly.
This separates two different questions: how much the API usage cost at the applicable rates, and how that amount was funded or billed. Anthropic’s pricing documentation describes its basis as “Billing is based on actual monthly usage.”
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.

