No plan or monthly fee: each request is billed at its model's rates. Your bill depends on how many tokens your prompts and answers use, so this page shows the rates and no monthly total.
Claude API list prices per 1M tokens, by model
Model
Input
Cache read
Cache write
Output
Claude Fable 5.1
$10.00
$0.25
$12.50 (5 min)$20.00 (1 h)
$50.00
Claude Opus 5.5
$4.00
$0.20
$5.00 (5 min)$8.00 (1 h)
$20.00
Claude Sonnet 5.5
$2.00
$0.10
$2.50 (5 min)$4.00 (1 h)
$10.00
Claude Haiku 5.5Prompts up to 100K tokens
$0.10
$0.01
$0.125 (5 min)$0.20 (1 h)
$0.50
Claude Haiku 5.5Prompts over 100K tokens
$0.50
$0.05
$0.625 (5 min)$1.00 (1 h)
$2.50
USD per 1M tokens. Standard processing. Cache write is the price of storing input in the prompt cache for the stated lifetime.
“Claude 4.7 and later models and Claude Mythos Preview use a newer tokenizer that produces approximately 30 percent more tokens for the same text than earlier models.”
No subscription: you prepay credits, which expire 1 year after purchase and aren't refundable. Invoiced billing needs a sales agreement.
Usage tiers set monthly spend caps (Start $500, Build $1,000, Scale $200,000). Reaching the cap pauses the API until the 1st of the next month.
Extended thinking tokens bill as output tokens.
Batch processing costs 50% less than the rates shown. Opus fast mode costs $8 input / $40 output per 1M tokens, and US-only inference costs 1.1× the standard rates.
Once a Claude Haiku 5.5 prompt passes 100K tokens (cache reads and writes count), every token in that request bills at the higher long-prompt rates in the table.
Tool fees: web search $10 per 1k searches; web fetch costs only its tokens; code execution includes 1,550 free hours a month, then $0.05 per hour.