What is Claude Fable 5?
Claude Fable 5 is Anthropic's generally available frontier model announced alongside Claude Mythos 5. This page separates official facts from practical interpretation.
Short answer: Claude Fable 5 is the public Claude model identified in Anthropic documentation as claude-fable-5. Official model documentation lists a 1M-token context window, 128K maximum output, adaptive thinking always on, and pricing of $10 per million input tokens and $50 per million output tokens.
Why Fable 5 matters
Fable 5 matters because it compresses several high-value capabilities into one public model: a very large context window, long output, deeper reasoning behavior, and stronger fit for agentic coding and research workflows. That does not mean every product should default to it. It means teams should reserve it for tasks where the model can use the extra context and reasoning budget.
The safe way to describe it
Use official language where possible. Say that Anthropic announced Claude Fable 5 and Claude Mythos 5, that Fable 5 is generally available, and that the model appears in Anthropic's model documentation with the listed API ID, context size, output limit, and token prices. Avoid claiming unsupported private benchmark details unless you can cite the specific source.
Independent-site note: fable-5.ai is not Anthropic. It is a third-party guide that links back to official Anthropic pages for source verification.
Best-fit workloads
- Claude Code sessions that need repository-wide reasoning, migration planning, or test repair.
- Long document analysis where the 1M-token window changes the workflow.
- Research synthesis that benefits from fewer retrieval rounds and more context at once.
- High-stakes drafting where a second or third cheaper retry would be more expensive than using a stronger model first.
Weak-fit workloads
Fable 5 is a poor default for cheap, high-volume chat, routine classification, simple extraction, or any workload where a lower-cost Claude model already meets quality targets. The model-selection question should be economic: does the additional capability reduce total cost, risk, or human review enough to justify the higher token price?
Related pages
Read the model and pricing guide, the Claude Code workflow page, or the official source list.