Back to models
Anthropic logo

Anthropic

Claude Haiku 4.5

Anthropic’s faster lower-cost Claude tier for lightweight production tasks.

Overall score
87
fastbudgetclassification
Context window
200K tokens
Speed
Fast
Input pricing
$1 / 1M input tokens
Output pricing
$5 / 1M output tokens

Score breakdown

Capability84
Use-case fit87
Cost efficiency90
Speed93
Reliability89
Agent readiness87
Ecosystem90

Scores combine benchmark signals, product experience, and editorial weighting. Use them as a practical guide, not an absolute truth claim.

Best for

Agent automationResearch

Works with

Anthropic APIautomation workersclassification flows

Modalities

textimagefile

Sources & trust

Officially verified core fields
Official linkSummaryDescriptionModalitiesContext windowMax outputTool supportPricingPricing page

Editorial fields such as shortlist guidance, strengths, caveats, and scoring remain clearly separated from official provider data.

Last verified: Apr 9, 2026

Strengths

  • Useful for faster Claude-aligned production paths.
  • Good step-down tier under Sonnet for routed stacks.

Things to watch

  • Not the best option for deep reasoning or hard coding.
  • Many teams will still want Sonnet for their main interactive tier.

Best for