See how ChatGPT, Perplexity and Google AI Overviews describe you today

Free AI Visibility Audit
Surge45°
Anthropic model°

Claude Haiku 4.5

Claude Haiku 4.5 is Anthropic's fastest and cheapest Claude model, built to run near-frontier quality work at low latency and low cost. Anthropic positions it as matching the performance of Claude Sonnet 4 while costing and waiting far less than the larger Claude models.

AnthropicVerified 24/09/2026Released 15/10/2025

01 / decision

Decision snapshot

Each figure sits next to the middle of the field, so it can be read as dear or cheap, wide or narrow, rather than floating on its own. Compared against the 126 models we track, not against an absolute standard.

Capability

Not measured

No published benchmark scores yet

Input / 1M tokens

$1.00

field median $0.30

Output / 1M tokens

$5.00

field median $1.25

Context

200K

field median 524K

02 / overview

What Claude Haiku 4.5 is, and when to reach for it

Four questions, answered separately, because somebody arrives at one of them rather than at the top of the page.

What it is

The job it was built for, and the job it is not for.

Claude Haiku 4.5 is the small, fast tier of Anthropic's Claude 4.5 family, made for high-volume work where response time and unit cost matter more than squeezing out the last increment of reasoning. It handles text, images and files, so it suits classification, extraction, summarisation, support replies and agent steps that fire thousands of times a day. It is not the model to reach for when a task needs the deepest reasoning Anthropic sells, that is what the larger Claude models are there for.

When it arrived, and when to use it

Where it sits in its line, and when a sibling is the better pick.

Claude Haiku 4.5 first appeared in October 2025 as the speed-and-cost tier alongside the bigger Claude 4.5 models. Pick it when a job runs at volume, when latency is visible to a user, or when you are fanning out many small calls inside an agent loop. Pick a larger Claude instead for long multi-step reasoning, hard code work, or anything where a wrong answer is expensive to catch.

How you reach it

The API, the apps it powers, and what its limits let you do.

Claude Haiku 4.5 is reached through the Anthropic API and is served in the Claude apps, where it acts as the quick tier behind everyday chat. Its context window is large enough to hold long documents, full support threads or a sizeable codebase slice in a single call, and it can return long outputs, so drafting, rewriting and structured extraction all fit in one pass. Because it accepts images and files as well as text, you can put screenshots, scanned documents and PDFs straight into the same pipeline.

Why it matters

What changes because this exists, or why it does not.

Claude Haiku 4.5 makes it practical to run Claude-grade quality on work that previously had to be sent to a cheaper, weaker model or batched overnight. Teams building agents get a model they can call in a loop without watching the bill climb, and product teams get a chat tier that answers fast enough to feel live. The change is economic rather than novel, the same tasks, at a volume that now adds up.

03 / evidence

How much of this is verified

Split by category, so a strong number never hides a thin evidence base. Verified means we read it on the benchmark's own published results; a provider's claim about its own model is shown and labelled rather than dropped.

No published benchmark scores for this model yet.

We publish a score only where we can link the result to where it was published. Until a benchmark result for this model exists in a source we read, this section stays empty rather than being filled with a provider's marketing figure.

How we decide what counts as evidence

04 / ledger

Benchmark ledger

Every published row, grouped by category, each compared with the best published score on the same benchmark. 'Is 64% good' is a question nobody can answer; '26 points behind the leader' is one anybody can.

Nothing in the ledger yet.

Each row here carries a score, the benchmark it came from, what the leading model scored on the same test, and a link to the published result. Rows appear as results are published and read.

How we decide what counts as evidence

05 / capability

Capability shape

Where this model is strong, and against how many peers. Ranks are against models with evidence in that category, not against everything we track: ranking against models nobody tested would rank who published, not who is better.

No category scores to shape yet.

A category score is the weighted mean of the benchmarks published for it. With no published rows there is nothing to average, and an empty chart drawn at zero would say something false.

How we decide what counts as evidence

06 / cost

What it costs

List API rates as last read from the provider, with the source on every row, plus every change we have recorded since we started tracking it.

Claude Haiku 4.5 API pricingSurge45°
ChargePriceUnitRead onSource
Input$1.00per 1M tokens2026-09-24Check
Output$5.00per 1M tokens2026-09-24Check

Claude Haiku 4.5 sits at the budget end of Anthropic's range, with output charged at several times the input rate as usual, so read-heavy tasks like classification and extraction are the cheapest thing you can run on it. Against the wider field it is priced as a small model rather than a frontier one, which is the point, it is meant to be called often enough that the per-call cost is the deciding factor.

What a month costsSurge45°
WorkloadInput / monthOutput / monthCost
A small product team20M tokens5M tokens$45.00
A busy support assistant200M tokens40M tokens$400.00
A document pipeline1000M tokens100M tokens$1,500.00

List API rates, no caching and no batch discount, which both providers offer and which change the answer a great deal. Treat these as the ceiling, not the bill.

07 / specs

Specifications

As listed by the provider's own catalogue and re-read every few hours. Anything absent is absent there too.

SpecificationSurge45°
Context window200,000 tokens
Maximum output64,000 tokens
Modalitiestext, image, file
Released15/10/2025
StatusCurrent
Catalogue identifieranthropic/claude-haiku-4.5

08 / lineage

Lineage

What this model replaced, what replaced it, and what else its provider has in the field.

Also from Anthropic

09 / line

The line

Every model in this family in release order, so a page from eight months ago says in one glance that two newer ones exist.

Nothing else in this line yet.

A line is worked out from the naming and the release dates across every model page we hold. It fills in as the provider ships successors, or as we pick up the models that came before this one.

How we decide what counts as evidence

10 / notes

Our notes

What this model changes for a brand trying to be cited in AI answers, and every change we have logged since it launched.

What it changes for you

A cheap, fast Claude tier means more products can afford to run an AI answer layer over their own content, so the number of surfaces summarising your category goes up, not down. Those summaries are usually built from retrieved pages rather than model memory, which puts the weight back on whether your documentation, pricing and comparison pages are clean, current and easy to quote. If your product only exists as claims on a marketing page, a model working at this speed and price will skip past you to a source it can lift a sentence from.

Where buyers meet this model

Buyers meet Claude Haiku 4.5 most often without naming it, as the fast tier answering inside the Claude consumer and mobile apps. Developers meet it directly through the Anthropic API and the major cloud model catalogues that resell Claude. It also turns up inside third-party products that use Claude for search, summarisation and support, where it does the high-volume retrieval and drafting work behind the scenes.

Change log

Nothing published here yet. Changes appear within hours of a provider announcing them.

11 / questions

Questions

The things people ask about this model, answered from what is on this page rather than from anywhere else.

Is Claude Haiku 4.5 good enough to replace a larger Claude model?
For high-volume, well-scoped work, often yes. Anthropic says it matches Claude Sonnet 4's performance at lower cost and latency. For long multi-step reasoning or tasks where an error is expensive, stay on a larger Claude.
Can Claude Haiku 4.5 read images and documents?
Yes. It accepts text, images and files, so screenshots, scans and PDFs can go through the same pipeline as plain text.
What is Claude Haiku 4.5 best used for?
Work that runs at volume or in front of a user: classification, extraction, summarisation, support replies, and the small repeated calls inside an agent loop.
Where can I use Claude Haiku 4.5?
Through the Anthropic API, inside the Claude apps as the fast tier, and in third-party products that build on Claude.
Surge45°

Is Claude Haiku 4.5 recommending you?

Models change what gets cited. We measure whether AI answers name your brand or your competitors across every assistant, and show you what to change.

Free AI visibility audit Book a call