Claude Haiku 4.5 API

Overview

Claude Haiku 4.5 is built for speed and volume. Classification, extraction, simple Q&A, content moderation: tasks where you need a fast, cheap answer and don't need flagship reasoning. Responses typically land in under a second.

What Claude Haiku 4.5 is best for

• High-volume classification and tagging
• Content moderation pipelines
• Simple Q&A and routing
• Latency-sensitive real-time features

Strengths and tradeoffs

Strengths: Lowest cost in the Claude family, fastest responses, still 200K context.

Tradeoff: Not built for deep reasoning or complex multi-step work. Use Sonnet or Opus for that.

Pricing

Estimate your own usage with the API cost calculator or count tokens with the token counter.

Quick start (Python)

Claude Haiku 4.5 runs through KissAPI's OpenAI-compatible endpoint, so any OpenAI SDK works. Just change the base URL and model id:

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_KISSAPI_KEY",
    base_url="https://api.kissapi.ai/v1",
)

resp = client.chat.completions.create(
    model="claude-haiku-4-5",
    messages=[{"role": "user", "content": "Explain async/await in one paragraph."}],
)
print(resp.choices[0].message.content)

Works with Cursor, Claude Code, Cline, Aider, and any OpenAI-compatible client.

Frequently asked questions

Related