Skip to content
LLM Toolkit

Model profile · Anthropic

Claude Haiku 4.5 pricing

Claude Haiku 4.5 costs $1.00 per 1M input tokens and $5.00 per 1M output tokens, with a 200K context window. Rates verified against the official Anthropic pricing page as of 2026-09-02.

Input

$1.00 /1M in

cached: $0.10

Output

$5.00 /1M out

5.0× the input rate

Context window

200K tokens

≈ 150,000 words

Claude Haiku 4.5 at a glance

Input / 1M tokens $1.00
Cached input / 1M $0.10
Output / 1M tokens $5.00
Context window 200K tokens
≈ English words of context 150,000
Provider Anthropic
Price verified 2026-09-02

Source: Anthropic official pricing page. We re-verify every model monthly.

What Claude Haiku 4.5 costs per month

Workload Claude Haiku 4.5 monthly cost Per request
Side-project chatbot 30K req · 2,000 in · 500 out · 40% cached $113.40/mo $0.0038
Coding agent 500K req · 12,000 in · 1,500 out · 70% cached $5,970/mo $0.012
Bulk extraction 2,000K req · 800 in · 120 out · 20% cached $2,512/mo $0.0013

Computed from list prices verified 2026-09-02. Cache savings blend normal and cached input rates by the hit rate. Compare against every other model at your exact volumes on the API Cost Calculator.

How much context is 200K tokens?

About 150,000 English words — roughly 300 pages of text — or several large codebases. Filling the entire window once costs $0.200 in input tokens at list price, before the model generates a single output token. Count your real documents with the AI Token Counter.

Frequently asked questions

How much does Claude Haiku 4.5 cost per million tokens?

Claude Haiku 4.5 costs $1.00 per 1M input tokens and $5.00 per 1M output tokens, with cached input at $0.10. Rates are verified against Anthropic's official pricing page as of 2026-09-02.

How big is Claude Haiku 4.5's context window?

Claude Haiku 4.5 ships a 200K token context window — roughly 150,000 English words, or about 300 pages of text. Filling the entire window once costs $0.200 in input tokens at list price.

Is Claude Haiku 4.5 cheap or expensive compared to other models?

Among the 22 models we track, Claude Haiku 4.5 ranks #9 on input price — cheapest is GPT-5 nano at $0.05, priciest is Claude Fable 5 at $10.00. At $1.00 per 1M input, Claude Haiku 4.5 runs 20.0× the cheapest tier. Compare it head-to-head with its rivals on the model comparison pages.

What is cached input pricing on Claude Haiku 4.5?

Claude Haiku 4.5 charges $0.10 per 1M cached input tokens — 10% of its $1.00 base input rate. If your requests repeat context (system prompts, documents, conversation history), prompt caching cuts the input portion of your bill accordingly. Model the exact savings on the API Cost Calculator.

More Anthropic models

Or see all rates on the Pricing Comparison.