Anthropic · current

Claude Haiku 4.5 API pricing

Economy Claude model for latency-sensitive and high-volume tasks.

Input / million
$1.00
Cached input / million
$0.1000
Output / million
$5.00
Context window
200,000

Example cost

A request with 10,000 input tokens and 2,000 output tokens costs approximately $0.0200 at the recorded standard rate.

  • 10,000 standard input tokens / 1,000,000 × $1
  • 0 cached tokens / 1,000,000 × $0.1
  • 2,000 output tokens / 1,000,000 × $5

Pricing rules and limitations

Fastest current Claude tier; cache reads and writes use distinct rates.

Price history

EffectiveUntilInputOutputStatus
2025-10-01Current / scheduled$1.00$5.00current