Skip to main content
Claude Sonnet 5 is Anthropic’s latest Sonnet model, available through Anyfast via the native Anthropic Messages API. It combines Sonnet-class speed with stronger coding, agentic execution, long-context reasoning, and analysis capabilities, making it well suited for production assistants, developer tools, and complex multi-step workflows.

​
Key capabilities

  • 1M context window — 1M tokens by default, with 128K max output tokens
  • Adaptive thinking — On by default; control depth with output_config.effort, or use thinking: {"type": "disabled"} to turn it off
  • Same API shape — Requests, responses, and streaming keep the same shape as Claude Sonnet 4.6
  • New tokenizer — The same text produces about 30% more tokens than Claude Sonnet 4.6
  • Model ID — Use claude-sonnet-5

​
Quick example

curl https://www.anyfast.ai/v1/messages \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5",
    "max_tokens": 1024,
    "messages": [
      { "role": "user", "content": "Write a Python function to merge two sorted arrays." }
    ]
  }'

​
Parameters

ParameterTypeRequiredDescription
modelstringYesMust be claude-sonnet-5
messagesarrayYesList of { role, content } objects
max_tokensintegerYesMaximum tokens to generate. Claude Sonnet 5 supports up to 128K output tokens.
output_configobjectNoUse {"effort":"low" | "medium" | "high" | "xhigh" | "max"} to control adaptive thinking depth. Default: high
thinkingobjectNoOmit it to use default adaptive thinking, or use {"type":"disabled"} to turn thinking off.
streambooleanNoEnable SSE streaming. Default: false
stop_sequencesarrayNoSequences that stop generation
Claude Sonnet 5 uses adaptive thinking by default. Use output_config.effort to tune reasoning depth (low, medium, high, xhigh, or max). Manual extended thinking (thinking: {type: "enabled", budget_tokens: N}) returns a 400 error, and non-default temperature, top_p, or top_k values also return a 400 error. Use thinking: {type: "disabled"} to turn thinking off.
Claude Sonnet 5 uses a new tokenizer. The same text produces approximately 30% more tokens than on Claude Sonnet 4.6, so prompt counts and max_tokens budgets should be recalculated before migrating.
Requests involving prohibited or high-risk cybersecurity topics may be refused. Refusals return stop_reason: "refusal".

API Reference

View the interactive API playground for Claude Sonnet 5.