Anthropic

Claude Haiku 3.5 API Pricing

Listed on the pricing page as "retired, except on Bedrock and Google Cloud." Exact apiId/context window/max output not confirmed this session. Cache write shown is 5m-TTL; 1h-TTL is $1.60/MTok. Batch: $0.40/$2.00 per MTok.

Could not verify pricing on 2026-07-02 — treat these numbers as tentative.

View source ↗

Facts

DeprecatedUnverified

Specs

  • FamilyClaude Haiku
  • API ID
  • ModalityMultimodal
  • Context window
  • Max output
  • Release dateNot published

Pricing (per 1M tokens)

  • Input$0.80 / 1M tokens
  • Output$4.00 / 1M tokens
  • Cached input$0.08 / 1M tokens
  • Cache write$1.00 / 1M tokens
  • Batch discount50%

Capabilities

vision, tool-use

Notes

Listed on the pricing page as "retired, except on Bedrock and Google Cloud." Exact apiId/context window/max output not confirmed this session. Cache write shown is 5m-TTL; 1h-TTL is $1.60/MTok. Batch: $0.40/$2.00 per MTok.

Worked cost examples

Estimated monthly cost on three common usage shapes

WorkloadEst. monthly cost

Chatbot

800 input / 200 output tokens per request, 100,000 requests/month

$144.00

RAG pipeline

6,000 input / 700 output tokens per request, 30,000 requests/month

$228.00

Batch summarization

20,000 input / 1,500 output tokens per request, 5,000 requests/month

$110.00

Price history

Changes we have documented since we began tracking

No price changes recorded since we began tracking.

Claude Haiku 3.5 pricing FAQ

How much does Claude Haiku 3.5 cost per 1M tokens?
Claude Haiku 3.5 costs $0.80 per 1M input tokens and $4.00 per 1M output tokens.
How much does Claude Haiku 3.5 cost per request?
A typical request of 800 input and 200 output tokens costs about $0.0014 on Claude Haiku 3.5.
Is Claude Haiku 3.5 cheaper than Claude Haiku 4.5?
Claude Haiku 3.5 is cheaper than Claude Haiku 4.5 on input tokens, by about 20%.
Does Claude Haiku 3.5 have cache pricing?
Yes — cached input reads cost $0.08 per 1M tokens, and cache writes cost $1.00 per 1M tokens.