Claude API Pricing Calculator

Estimate your monthly Claude API bill across Haiku, Sonnet, Opus, and Fable — with caching and Batch API discounts built in.

💡 Uses Anthropic's published per-token rates as of July 2026. Always confirm current prices on Anthropic's official pricing page before budgeting.

Estimate Your Monthly Cost

Model

Requests / month

Avg. input tokens / request

Avg. output tokens / request

Cached input share 0%

Use Batch API (50% off, async processing)

$0.00estimated / month

Input tokens (monthly)—

Output tokens (monthly)—

Input cost—

Output cost—

Cost per request—

Total / month—

Model	Rate ($/MTok)	Est. Monthly Cost

About

What is this Tool?

The Claude API Pricing Calculator helps developers and teams estimate what a workload will actually cost before it hits production. Pick a model, enter your monthly request volume and average token counts, and toggle prompt caching or Batch API discounts to see how they change your bill. It also compares your workload across every current Claude model tier — Haiku 4.5, Sonnet 5, Opus 4.8, and Fable 5 — side by side, so you can decide whether a cheaper or more capable model makes sense for what you're building.

How It Works

Estimate Your Bill in 4 Steps

No spreadsheet needed. Fill, calculate, compare.

Choose Your Model

Select the Claude model you're using or considering — Haiku 4.5, Sonnet 5, Opus 4.8, or Fable 5 — each with its own per-token rate.

Enter Your Volume

Add your expected monthly request count, plus average input and output tokens per request. Rough numbers from your logs or a test run work fine.

Apply Discounts

Slide in your cached input share if you reuse system prompts or documents, and toggle Batch API if your workload can run asynchronously for 50% off.

Calculate & Compare

Click Calculate to see your estimated monthly cost, then switch to the comparison tab to see the same workload priced across every model tier.

Reference

Current Claude API Rates

Standard per-token pricing this calculator uses, for prompts under 200K tokens.

Model	Input ($/MTok)	Output ($/MTok)	Best For
Claude Haiku 4.5	$1.00	$5.00	High-volume, simple tasks
Claude Sonnet 5	$3.00$2 thru Aug 31	$15.00$10 thru Aug 31	Coding & agentic workflows
Claude Opus 4.8	$5.00	$25.00	Hardest reasoning tasks
Claude Fable 5	$10.00	$50.00	Specialized Mythos-tier work

Prompt caching reads cost roughly 10% of the standard input rate. Batch API cuts both input and output rates in half for asynchronous jobs. Prompts over 200K tokens carry a long-context surcharge on Opus and Sonnet. Rates change over time — check claude.com/pricing for the current numbers.

Is this calculator free to use?

Yes, completely free. No sign-up required. Enter your numbers and get an instant estimate as many times as you like.

Where do the prices come from?

Rates reflect Anthropic's published per-token API pricing as of July 2026, including Claude Sonnet 5's introductory rate through August 31, 2026. Anthropic can change pricing at any time, so treat this as an estimate and confirm against the official pricing page before committing budget.

What does "cached input share" mean?

If you reuse the same system prompt, tool definitions, or document across requests, prompt caching lets you read that repeated content at a fraction of the standard input price. Set this to the rough percentage of your input tokens that come from repeated, cacheable content.

What is the Batch API discount?

The Batch API processes requests asynchronously, typically within 24 hours, at half the standard input and output price. It's a good fit for non-urgent workloads like bulk classification, evaluations, or nightly document processing.

Does this account for long-context surcharges?

No — this calculator uses standard rates for prompts under 200K tokens. If your requests regularly exceed that, Opus and Sonnet pricing rises for the portion above the threshold, so your real bill may be higher than the estimate shown here.

Important Note

Before You Budget

A quick reminder about using this estimate.

⚠️

Disclaimer: This calculator provides estimates based on publicly listed per-token rates and simplified assumptions about caching and batch usage. It does not account for every pricing variable (long-context surcharges, regional multipliers, tool-specific fees, or rate changes). Always confirm current pricing on Anthropic's official pricing page before finalizing a budget. ToolsRevis is not responsible for discrepancies between this estimate and your actual invoice.