Provider pricing guide

Anthropic Claude API Pricing Calculator & Model Cost Guide

Claude model tiers separate standard input, cache-hit input, output, and batch rates. Output-heavy agent and coding workloads can cost substantially more than input-only comparisons suggest.

Open Claude cost calculator

ProviderModality

Example comparison basis: 1M input tokens + 1M output tokens. This is a rate comparison, not a quality or capability ranking.

AI model pricing comparison filtered by provider and modality
Provider / model	Input	Output	Cached / batch	Unit price	Example cost	Verified / source	Use
Anthropic Claude Haiku 4.5 Cheapest	$1.00 / 1M	$5.00 / 1M	$0.10 cache	—	$6.00	Verified Jun 21, 2026 Anthropic Claude API pricing	Calculate
Anthropic Claude Sonnet 4.6	$3.00 / 1M	$15.00 / 1M	$0.30 cache	—	$18.00	Verified Jun 21, 2026 Anthropic Claude API pricing	Calculate
Anthropic Claude Opus 4.8 Highest	$5.00 / 1M	$25.00 / 1M	$0.50 cache	—	$30.00	Verified Jun 21, 2026 Anthropic Claude API pricing	Calculate

Pricing FAQ

How do I calculate Claude API cost?

Divide input and output token counts by one million, multiply by their respective Claude model rates, add the results, and multiply by request volume.

What is the cost difference between Claude Sonnet and Opus?

The exact difference depends on the input-to-output ratio. Use the model comparison or calculator with the same workload to avoid a misleading unit-rate comparison.

Does prompt caching reduce Claude cost?

Eligible cache hits use a separate lower rate, while cache writes can have distinct rates. Verify Anthropic's current cache rules before forecasting savings.

Related comparisons

Claude Sonnet vs Opus API Cost Comparison

Compare Claude Sonnet vs Opus API cost using source-linked pricing and a consistent monthly request workload.

Open

OpenAI vs Anthropic API Cost Comparison

Compare representative OpenAI and Anthropic API model costs for the same text workload and monthly request volume.

Open

Related glossary terms

Input tokens

Input tokens are the tokenized units sent to a model, including instructions, user content, conversation history, retrieved context, and tool definitions.

Open

Output tokens

Output tokens are the tokenized units generated by a language model, including visible responses and any billable reasoning or thinking tokens defined by the provider.

Open

Cached tokens

Cached tokens are repeated prompt tokens that qualify for a provider's prompt-caching mechanism and may use a separate cache-hit rate.

Open