How large is Claude context window?

All Claude 3 models (Sonnet, Opus, Haiku) support a 200,000-token context window — equivalent to about 150,000 words or roughly 400 pages of text. This is larger than GPT-4o but smaller than Gemini 2.5 Pro.

Can Claude use its full up to 1M context effectively?

Claude performs best on retrieval tasks within its context window, though recall quality can degrade with very long documents. For best results, place the most important content at the beginning or end of the context.

Does Claude count system prompt tokens differently?

Anthropic charges for system prompt tokens the same as user message tokens (input rate). Claude Sonnet 4.6 is $3 per million input tokens including system prompt tokens.

Claude Context Window Calculator

Calculate token usage against Claude up to 1M context windows.

Check whether your system prompt, user message, and expected output fit within Claude Sonnet 4.6, Claude Opus 4.6, or Claude Haiku 4.5 context windows. All Claude models support 200,000 tokens — about 150,000 words.

Model

System prompt

User prompt

Expected output tokensMax output for Claude Sonnet 4.6: 16,000 tokens

Total context usage

5000.3% of 200.0K

System tokens0

User tokens0

Output tokens500

Remaining199,500

Related Tools

CLTClaude Token CounterNEW

Count tokens for Claude Sonnet 4.6, Opus, and Haiku models.

CLDClaude API Cost CalculatorNEW

Calculate Anthropic Claude API costs for Sonnet, Opus, and Haiku.

CTXAI Context Window CalculatorNEW

Check if your prompts fit within any AI model context window.

GPTOpenAI Context Window CalculatorNEW

Check if your prompts fit within GPT-4o and GPT-3.5 context windows.

FAQ

How large is Claude context window?: All Claude 3 models (Sonnet, Opus, Haiku) support a 200,000-token context window — equivalent to about 150,000 words or roughly 400 pages of text. This is larger than GPT-4o but smaller than Gemini 2.5 Pro.
Can Claude use its full up to 1M context effectively?: Claude performs best on retrieval tasks within its context window, though recall quality can degrade with very long documents. For best results, place the most important content at the beginning or end of the context.
Does Claude count system prompt tokens differently?: Anthropic charges for system prompt tokens the same as user message tokens (input rate). Claude Sonnet 4.6 is $3 per million input tokens including system prompt tokens.