Batch summaries
100k input tokens and 50k output tokens for a content batch.
Model calculator
Claude Token Calculator
Convert Claude token counts into cost and compare Opus, Sonnet, and Haiku for the same job.
Pricing table last updated: 2026-05-13
Interactive tool
Convert tokens to cost
Use presets, share the exact inputs, and scan the live breakdown.
Estimated total cost$1.05
Cost per 1k tokens$0.007
Examples
Short support reply
2,000 prompt tokens and a compact 300-token answer.
Claude Sonnet 4.6: $3.00 input / $15.00 output per 1M tokens.
How this Claude Token Calculator works
How Claude counts tokens
Claude uses Anthropic's own tokenizer, so the same text often tokenizes to a slightly different count than GPT — don't assume a GPT estimate transfers exactly. Roughly three-and-a-half to four characters per token works for rough English planning, and Anthropic's token-counting endpoint gives exact figures. The system prompt and prior conversation are resent on every turn, so input grows as a session lengthens.
Choosing the right Claude model
The lineup is tiered: Claude Opus 4.7 for the most demanding reasoning and agentic work (highest price), Claude Sonnet 4.6 as the balanced production default, and Claude Haiku 4.5 for fast, inexpensive, high-volume tasks. Output is priced above input on each tier, so generation-heavy features lean on output cost — switch models in the calculator above to compare the same workload.
Prompt caching and what's excluded
Anthropic's distinctive lever is prompt caching: reusing a large, stable prefix such as a long system prompt or reference document sharply cuts the input cost of repeated calls. This calculator uses standard uncached pricing, so caching-heavy workloads cost less than shown. It also excludes the Batch API, long-context premiums, and tool use — verify the real numbers in the Anthropic console.
Examples
Claude Sonnet 4.6 · 2,000 input / 500 output per reply. Balanced quality and price for production chat.
Claude Haiku 4.5 · many short messages tagged and routed. Cheapest tier for simple, repetitive calls.
Claude Opus 4.7 · long shared context, detailed output. Prompt-cache the context to cut repeated input.
FAQ
Are these AI costs exact?
They are estimates based on public token prices. Your bill can change with cached tokens, batch discounts, image or audio usage, taxes, provider credits, and model-specific rules.
Which currency does the calculator use?
All calculators use USD by default because major AI providers publish API pricing in USD.
Do Claude and GPT count tokens the same way?
No. Anthropic and OpenAI use different tokenizers, so the same text can produce different token counts. Estimate Claude costs with Claude token counts rather than reusing a GPT figure.
How does prompt caching change the cost?
Anthropic can cache a stable prompt prefix so repeated requests reread it cheaply. For workloads that resend a large system prompt or document, real input cost can fall well below this standard-pricing estimate.
Which Claude model should I default to?
Sonnet is the common production balance of quality and price. Move up to Opus for genuinely hard reasoning, or down to Haiku for cheap, high-volume tasks.