Skip to main content
Updated: September 2, 2026 BetterToken now supports Claude Fable 5.1 with the exact Model ID claude-fable-5-1. Anthropic positions it for demanding reasoning and long-horizon agentic work, including complex coding, multistep research, and document-heavy workflows. Check current availability and BetterToken pricing in the model plaza.

​
Highlights

  • Long-running agentic work: designed for complex tasks that can span many steps, tools, and applications
  • Coding and knowledge work: optimized for ambitious codebase changes, code review, research, and work with documents, spreadsheets, slides, charts, and diagrams
  • Large context and output: Anthropic lists a 1M-token context window and up to 128K output tokens
  • Adaptive thinking: always enabled, with high as the default effort level in the Claude API
  • Lower cache-read cost: Anthropic prices cache reads at one quarter of the Claude Fable 5 rate; BetterToken prices may differ, so use the model plaza for current rates

​
Configuration recommendations

  • Use Fable 5.1 for demanding reasoning or long-running work. Anthropic recommends starting with Claude Opus 5 for most workloads and moving to Fable 5.1 when your own evaluations justify it.
  • Test the model with a representative task before switching production traffic. Compare output quality, latency, and token usage in your own client.
  • If you migrate from Claude Fable 5, keep conversation history append-only. Fable 5.1 rejects forced tool use, earlier models cannot read its thinking blocks, and editing earlier turns invalidates those blocks.
  • Always use the exact BetterToken Model ID claude-fable-5-1. Do not substitute a display name or a platform-specific ID.

​
Quick start

  1. Create a BetterToken API Key or open the Setup dialog for an existing key.
  2. Confirm that claude-fable-5-1 is available in the model plaza.
  3. Open the setup guide for your client and enter the BetterToken API Key, required Base URL, and exact Model ID.
  4. Restart the client if it does not refresh its model list automatically.
  5. Send a short test request and verify it in the Dashboard.
Use one of these guides:

​
Frequently asked questions

​
When should I use Claude Fable 5.1?

Choose it for demanding reasoning, long-running coding agents, multistep research, or complex document work. For routine workloads, compare it with Claude Opus 5 and other available models before deciding.

​
Does BetterToken provide the full 1M-token context window?

Anthropic lists a 1M-token context window for Claude Fable 5.1. The usable limit can also depend on your client and request settings. Check the model plaza and your client documentation before sending very large prompts.

​
How do I fix Model not found?

Confirm that the Model ID is exactly claude-fable-5-1, the BetterToken API Key can access the model, and the Base URL matches your client’s setup guide. Restart the client and test again.

​
Can I replace Claude Fable 5 with Fable 5.1 directly?

Start by changing the Model ID and running your existing evaluations. Existing prompts should usually remain useful, but review forced tool use, cross-model thinking blocks, and any code that edits earlier conversation turns.
Availability and BetterToken pricing can change. Check the live model plaza before configuring the model.

​
Official sources