Claude Haiku 5.5

Claude Haiku 5.5 (anthropic/claude-haiku-5.5) is the cheapest fifth-generation Claude: classification, extraction, drafts, bulk processing. It accepts images, reasons and has native web search — rare in this price class.

Parameters and capabilities

Parameter	Value
model	anthropic/claude-haiku-5.5
Images on input	yes (image_url, data URLs)
Files on input	no
Reasoning	reasoning_effort: low, medium, high, xhigh, max; disabled with off
Native web search	yes — web_search: "native"
Cache read	$0.012 per 1M tokens

Example

Python

from openai import OpenAI

client = OpenAI(base_url="https://api.mixen.ai/v1", api_key="mxn-...")

resp = client.chat.completions.create(
    model="anthropic/claude-haiku-5.5",
    messages=[{"role": "user", "content": "Tag this support ticket with a topic and a category"}],
    reasoning_effort="low",
)
print(resp.choices[0].message.content)

cURL

curl https://api.mixen.ai/v1/chat/completions \
  -H "Authorization: Bearer $MIXEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-haiku-5.5",
    "messages": [{"role": "user", "content": "Tag this support ticket with a topic and a category"}]
  }'

The Anthropic protocol works with the same key: base URL https://api.mixen.ai, the SDK appends /v1/messages itself.

Pricing

Per 1M tokens: input — $0.12, output — $0.6. Cache: read — $0.012. Showcase prices follow the live exchange rate — take fresh values from the catalog.

Strengths and limitations

• Cheap classification and extraction: the full set of reasoning levels, including off for pure speed.
• Cache read at $0.012 — long sessions on repeating prefixes cost pennies.
• Native web search is available — but watch the per-request charge on bulk jobs.
• No file input: extract PDF/DOCX to text beforehand.

For harder jobs — Sonnet 5.5 and Opus 5.5. All models — in the catalog.