Compare

Claude Haiku 4.5 vs Claude Sonnet 5

Same two models, priced on three workload shapes. The rate card rarely decides this on its own — the ratio of input to output tokens usually does.

Per million tokens

Claude Haiku 4.5

Claude Sonnet 5

Input$1.00$2.00
Output$5.00$10.00
Cached input$0.10$0.30
Context200K1M
Max output64K128K
ProviderAnthropicAnthropic

Monthly cost by workload

Chat assistant

60,000 requests · 1,200 in / 400 out · 30% cached — short prompts, short answers, high volume

Claude Haiku 4.5

$172.56

Claude Sonnet 5

$347.28

Claude Haiku 4.5 is 2.0× cheaper on this shape — $174.72 a month.

RAG over documents

20,000 requests · 18,000 in / 700 out · 75% cached — large retrieved context, heavily cached prefix

Claude Haiku 4.5

$187.00

Claude Sonnet 5

$401.00

Claude Haiku 4.5 is 2.1× cheaper on this shape — $214.00 a month.

Coding agent

4,000 requests · 45,000 in / 6,000 out · 80% cached — long tool-calling loops, big cached prefix

Claude Haiku 4.5

$170.40

Claude Sonnet 5

$355.20

Claude Haiku 4.5 is 2.1× cheaper on this shape — $184.80 a month.

Run these numbers on your own workload