Models & pricing / claude-sonnet-5-5 · Anthropic
Anthropic

claude-sonnet-5-5 API

Claude Sonnet 5.5 is an Anthropic model for everyday coding, agent tool use and enterprise knowledge work, including well-scoped features, bug fixes and analysis. Check current XiuAI prices and protocols and choose thinking settings for the task.[1][2][4]

claude-sonnet-5-5

Model overview

Published provider specifications. Available features depend on the current API route.

Input
text / image
Output
Text
Context
1M
API routes
OpenAI-compatible API / openai-response / openai-response-compact / Anthropic Messages / Gemini GenerateContent / openai-alpha-search

Well-scoped everyday tasks

Consider feature development, bug fixes, data analysis and content work with defined inputs and acceptance criteria. State required behaviors, output formats and checks, and have the application validate the result. Tool use alone is not a reason to run an unclear goal without review.[1]

For work requiring longer autonomous execution or more complex systems judgment, compare Opus 5.5 or a current Fable version. Use the same tasks and current costs; family positioning does not guarantee a faster or cheaper result on every request.[1][4]

Thinking settings and migration limits

Adaptive thinking is on by default. Anthropic provides between_tools to turn off up-front thinking at high effort or below; that is not the older thinking.type=disabled setting. Do not carry over temperature, top_p, top_k or forced tool choice from older requests.[1][3]

API access

Access and cost checks

Use claude-sonnet-5-5 with your existing XiuAI key and a protocol shown on this page. Handle tool parameters and thinking for the target protocol and read content blocks by type. XiuAI provides model access; your application executes tools, stores data and performs final checks.[4][3]

Repeated material in a conversation affects cache-read and cache-write charges. Compare current cache conditions and actual usage when moving from Sonnet 5 to 5.5, rather than treating a vendor cache-price change as a fixed reduction for every XiuAI task.[3][4]

Send a request using your existing API key.

POST /v1/chat/completions
curl 'https://router-api.xiu.ai/v1/chat/completions' \
  -H "Authorization: Bearer $XIUAI_API_KEY" \
  -H 'Content-Type: application/json' \
  --data-raw '{
  "model": "claude-sonnet-5-5",
  "messages": [
    {
      "role": "user",
      "content": "Hello"
    }
  ]
}'
Read the API documentation

Frequently asked questions

Does between_tools work at every effort level?

No. Current documentation requires high effort or below. Confirm that reducing up-front thinking is the goal, then use a supported setting for the target protocol rather than mixing disabled, older manual budgets and between_tools.[1][3]

Why can a request copied from an older Sonnet return 400?

Check manual thinking budgets, sampling parameters, assistant prefill, forced tool_choice and thinking.type=disabled. The migration guide lists these as settings that need adjustment; remove or replace them with supported equivalents.[3]

Does image understanding mean it generates images?

Sonnet 5.5 accepts text and images and directly produces text. It can analyze visual material or write code, but image generation requires an appropriate media model and API.[1][4]

Related models

claude-opus-5-5claude-haiku-5-5claude-sonnet-4-6

Sources and updates

Specifications reference provider documentation; prices and availability come from the current XiuAI catalog. Specifications checked: 2026-11-11

Content reviewed: 2026-11-16

[1] Claude Sonnet 5.5 model reference[2] Claude Sonnet 5.5 announcement[3] Anthropic migration guide[4] Current XiuAI model and pricing dataTool callingStructured outputContext documentation