Anthropic

Claude Context Cost Calculator

Calculate Anthropic Claude API costs for Claude Sonnet 4 and Claude Haiku 3.5. Both models offer a 200K context window with competitive quality for complex tasks.

LLM Context Cost Calculator

Enter your usage parameters below

Number of tokens in your prompt per request

Number of tokens in the model response

Average number of API calls per day

$2.5/M input · $10/M output · 128k context

Cost Per Request

$0.022

GPT-4o - OpenAI

Daily Cost

$2.25

Monthly Cost

$67.50

Annual Cost

$810.00

Cost Per 1K Tokens

$0.0037

Blended input + output rate

Input vs Output Split

56% / 44%

Input cost vs output cost

Cheapest Alternative

GPT-4.1-nano

$2.70/mo - save 96% ($64.80/mo)

OpenAI · budget

Optimize your LLM spend

Platforms like OpenRouter, Together AI, and Groq offer competitive pricing and unified APIs across multiple models. Fireworks AI and Anyscale specialize in high-throughput inference at lower cost.

OpenRouterTogether AIGroqFireworks AIAnyscale

Need help optimizing LLM costs?

Digital Signet builds AI-powered systems and provides fractional CTO leadership. 20+ years shipping software.

This costs you ~$810/year

We'll identify the top 3 drivers and give you a 90-day mitigation plan.

Get a Free Exposure Teardown →

Or email Oliver directly → [email protected]

Claude Model Pricing

ModelInput $/1MOutput $/1MContextCategory
Claude Sonnet 4$3.00$15.00200Kflagship
Claude Haiku 3.5$0.80$4.00200Kfast

Claude Prompt Caching

Claude's prompt caching charges cache writes at $3.75/1M and cache reads at $0.30/1M - a 90% discount on input tokens. Break-even after ~1.2x reuse. Ideal for long system prompts and repeated document context.

Frequently Asked Questions

How much does the Claude API cost?

Claude Sonnet 4 is priced at $3.00 per 1M input tokens and $15.00 per 1M output tokens. Claude Haiku 3.5 is considerably cheaper at $0.80/1M input and $4.00/1M output. Both models have a 200,000 token context window. For a typical exchange (1,000 input + 500 output tokens), Claude Sonnet 4 costs about $0.0105 per call.

What is Claude's context window size?

All current Claude models support a 200,000 token context window - equivalent to roughly 150,000 words or a full novel. This is larger than GPT-4o's 128K but smaller than GPT-4.1's 1M context. Claude's 200K window is sufficient for most enterprise use cases including long document analysis, large codebase review, and extended multi-turn conversations.

How does Claude compare to GPT-4 on cost?

Claude Sonnet 4 at $3.00/1M input is 20% more expensive than GPT-4.1 ($2.00/1M), with higher output costs ($15 vs $8/1M). Claude Haiku 3.5 at $0.80/$4.00 is more expensive than GPT-4.1-mini ($0.40/$1.60) but often preferred for its stronger instruction following and reduced hallucinations in structured output tasks.

Does Claude support prompt caching?

Yes. Anthropic offers prompt caching for Claude models. Cached input tokens cost $0.30/1M (90% discount vs standard input price). Cache writes cost $3.75/1M. This is highly effective for applications with long static context like system prompts, retrieved documents, or tool definitions that are reused across many API calls.