About Claude Haiku 4.5
Claude Haiku 4.5 is the speed tier of Anthropic's lineup — its documentation calls it the fastest model with near-frontier intelligence. It is the one current Claude with the smaller context window and output ceiling, and the one that uses explicitly enabled extended thinking rather than the adaptive thinking of its larger siblings — a deliberate trade that keeps latency at the bottom of the range.
Its home ground is the high-frequency layer of a product: routing, moderation, extraction, first-draft responses and other tasks where a larger Claude would spend most of its capability waiting on the next request.
Haiku is the cheapest way into the Claude family, and the gap widens with caching — cache reads price at a tenth of an already-low input rate, so a system-prompt-heavy service runs remarkably lean. Two limits shape its economics: the context window is a fraction of its siblings', so very long documents need chunking or a bigger model, and extended thinking must be switched on deliberately, billing as output when used. A common architecture pairs it with a larger Claude — Haiku triages and handles the routine majority, escalating the hard minority — which usually beats running everything one tier up.
Frequently asked questions
How much does the Claude Haiku 4.5 API cost?
Claude Haiku 4.5 costs $1 per million input tokens and $5 per million output tokens, with cached input at $0.1 per million (90% off).
How much does a typical chat message cost with Claude Haiku 4.5?
A short chat message (about 500 input and 250 output tokens) costs roughly $0.0018 with Claude Haiku 4.5. At 1,000 such requests a day, that is about $52.50 per month.
Is Claude Haiku 4.5 cheap compared to similar models?
Claude Haiku 4.5 has the #13 cheapest input rate of the 29 models we track. On a mixed workload (10K input / 2K output), it costs $0.020 per request, versus $0.015 for Gemini 3.8 Flash and $0.021 for DeepSeek V4 Pro.
What does Claude Haiku 4.5 cost per 1,000 tokens?
Per 1,000 tokens, Claude Haiku 4.5 costs $0.0010 for input and $0.0050 for output, or $0.00010 for cached input. Providers quote rates per million tokens ($1/M in, $5/M out here), so divide by a thousand for the per-1K figure older pricing pages used.
How large a prompt can Claude Haiku 4.5 take?
Claude Haiku 4.5 accepts up to 195K tokens of input context and can generate up to 63K tokens of output per request. At its input rate, filling the entire 195K-token window costs about $0.200 per request before any output.
Related
Compare all models at once ·Count tokens in your prompt
All Anthropic model costs: Claude Fable 5.1 · Claude Fable 5 · Claude Opus 5 · Claude Sonnet 5 · Claude Opus 4.8
Last updated 2026-09-03. Prices verified against Anthropic's official pricing page; see the methodology.