Anthropic: Claude Haiku 4.5 (latest)

Fast Claude lane for lightweight agents, office tasks, and responsive chat

Model lab: Anthropic
Model family: claude-haiku

Modalitiestext, image, pdf → text

Context window200,000

Maximum output64,000

InputUSD 0.7968 / 1M tokens

OutputUSD 3.9856 / 1M tokens

Try now

Quickstart

Use your Moduck API Key, set the API base URL to https://api.moduck.ai/v1, and set the model parameter to anthropic/claude-haiku-4-5.

import os
import requests

base_url = "https://api.moduck.ai/v1"
api_key = os.environ["MODUCK_API_KEY"]
model_id = "anthropic/claude-haiku-4-5"

response = requests.post(
    f"{base_url}/chat/completions",
    headers={
        "Authorization": f"Bearer {api_key}",
        "Content-Type": "application/json",
    },
    json={
        "model": model_id,
        "messages": [{"role": "user", "content": "Hello!"}],
        "stream": False,
    },
    timeout=60,
)
response.raise_for_status()
print(response.json())

Use the familiar OpenAI-style chat request format to make model integration easier.

Claude Haiku 4.5 (latest) API Pricing

Type	Price information
Input	USD 0.7968 / 1M tokens
Output	USD 3.9856 / 1M tokens
Cached input	USD 0.0792 / 1M tokens
Cache write	USD 0.9488 / 1M tokens

Benchmarks

Benchmark	Score	Measurement	Source
SWE-Bench Pro	39.45	resolve rate · Dataset: public

Performance

Performance trends

Throughput88.5 tok/sCurrent

Time to first token1.2 sCurrent

End-to-end latency7.1 sCurrent

Throughput

Current88.5 tok/s

Window average89.7 tok/s

Time to first token

Current1.2 s

Window average1.4 s

End-to-end latency

Current7.1 s

Window average7.5 s

Tail latency (P99)

Current13.3 s

Window average16.7 s

Prompt cache hit rate

Current19.3%

Window average24.5%

Error rate

Current0.0%

Window average0.1%

Availability

Uptime100%At least one provider responded

Availability100%Requests actually served

Error rate0%Share of failed requests

Served requests, hour by hour

24 hours agoNow

Claude Haiku 4.5 (latest) Compare models

Model	Input price	Output price	Context window	Capabilities
Claude Haiku 4.5 (latest)This page	USD 0.7968 / 1M tokens	USD 3.9856 / 1M tokens	200,000	Text output · Reasoning · Tool calling
Claude Haiku 5.5	USD 0.08 / 1M tokens	USD 0.4 / 1M tokens	1,000,000	Text output · Reasoning · Tool calling
Claude Haiku 4.5	USD 0.8 / 1M tokens	USD 4 / 1M tokens	200,000	Text output · Reasoning · Tool calling
Claude Haiku 3	USD 0.2 / 1M tokens	USD 1 / 1M tokens	200,000	Text output · Tool calling
Claude Sonnet 5.5	USD 1.6 / 1M tokens	USD 8 / 1M tokens	1,000,000	Text output · Reasoning · Tool calling

Model specifications

Input: text, image, pdf
Output: text
Context window: 200,000
Maximum output: 64,000
Release date: 2025-10-15
Attachments: Yes
Reasoning: Yes
Tool calling: Yes
Temperature setting: Yes
Open weights: No

Frequently asked questions

What is Claude Haiku 4.5 (latest) for?

Fast Claude lane for lightweight agents, office tasks, and responsive chat

What inputs and outputs does Claude Haiku 4.5 (latest) support?

Recorded inputs: text, images, PDF; outputs: text.

How is Claude Haiku 4.5 (latest) priced?

The pricing table shows the current recorded Moduck price. For example: Input USD 0.7968 / 1M tokens; Output USD 3.9856 / 1M tokens. See the pricing table for all units, variants, and conditions.

What specifications are recorded for Claude Haiku 4.5 (latest)?

Current model records list Context window 200,000 tokens; Maximum output 64,000 tokens.

What model parameter does the Quickstart use?

The example uses anthropic/claude-haiku-4-5. This value comes from the page identifier and is not confirmed as a callable Moduck public ID.

Can I call this model through the Moduck API now?

This page has no verified Moduck API call for this model. A Quickstart example or reserved ID does not establish access; check the models available to your account.