Models & pricing / claude-fable-5-1 · Anthropic
Anthropic

claude-fable-5-1 API

Claude Fable 5.1 is a Fable version for sustained coding, research and knowledge work, with always-on adaptive thinking. Its tool-choice and cache conditions differ from Fable 5. Choose it on XiuAI using current pricing and migration requirements.[1][2][4]

claude-fable-5-1

Model overview

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

API routes
OpenAI-compatible API / openai-response / openai-response-compact / Anthropic Messages / Gemini GenerateContent / openai-alpha-search

Research and sustained revision

Consider research across multiple sources, extended code refactoring or analysis that repeatedly checks hypotheses. The release emphasizes sustained work and root-cause investigation. Preserve sources and intermediate conclusions, and ask for verifiable next steps rather than treating a finished analysis as established fact.[2]

If a smaller model already meets the task’s requirements, compare costs under the same conditions before choosing Fable. Launch benchmarks and cache-savings percentages reflect vendor conditions, not a fixed gain for every XiuAI request.[2][4]

Tool choice and version switching

Native Fable 5.1 does not accept disabled thinking, manual budgets, assistant prefill or forced any/tool selection; sampling fields should also be omitted. Tool support does not imply every tool-choice mode is supported, and compatibility mappings need their own checks.[3]

Moving to 5.2 also requires checking edits to earlier history, thinking-block compatibility and tool versions. Anthropic states that 5.1 cannot read 5.2’s thinking blocks; a successful model switch does not mean all reasoning state was preserved.[3]

API access

Access, caching and data handling

Use claude-fable-5-1 with your existing XiuAI key. Costs follow the current input, output, cache-read and cache-write conditions on this page. Reconcile multi-turn request records for long tasks instead of considering only the first input.[4][1]

Current migration documentation specifies model-specific retention requirements, with express authorization needed for zero retention. Launch statements about eligible customers or future enterprise arrangements do not mean this integration defaults to zero retention. Check provider and XiuAI processing separately.[3][2]

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-fable-5-1",
  "messages": [
    {
      "role": "user",
      "content": "Hello"
    }
  ]
}'
Read the API documentation

Frequently asked questions

Why can Fable 5 tool settings be incompatible with 5.1?

Fable 5 accepts forced any or tool choices, but 5.1 restricts them under the same native rules as 5.2. Check the target model’s tool-choice and structured-output support, not just tool names.[3]

Will switching from 5.2 back to 5.1 preserve all reasoning state?

There is no such guarantee. Anthropic states that 5.1 cannot read 5.2 thinking blocks. Retained text and tool messages do not imply identical reasoning state. Validate behavior and costs after switches using the documented compatibility rules.[3]

Do launch-era enterprise privacy plans establish zero retention today?

No. Current model documentation and applicable authorization determine provider requirements. Future plans, particular customer eligibility and XiuAI’s own logging policy do not replace that confirmation.[2][3]

Related models

claude-fable-5-2claude-opus-5-5

Sources and updates

Specifications reference provider documentation; prices and availability come from the current XiuAI catalog.

Content reviewed: 2026-11-16

[1] Claude Fable 5.1 model reference[2] Claude Fable 5.1 announcement[3] Anthropic migration guide[4] Current XiuAI model and pricing dataModel documentationTool callingStructured outputContext documentation