Skip to main content

Softened — This model answers instead of refusing, but requests beyond its own limits come back toned down, even with a fiction-writing system prompt. The call still succeeds and is still billed, so read the reply rather than trusting the status code.Browse every model at this filtering level
anthropic/claude-fable-5/chat

Claude Fable 5 chat from Anthropic through an OpenAI-compatible endpoint. It takes up to 1,000,000 tokens of text and image context and returns up to 128,000, with three reasoning-effort levels, up to 128 tool definitions and SSE streaming.

Input $9Output $45per 1 million tokens10% below the publisher's list priceCached input $0.90Cache write $11Anthropic official priceInput $10Output $50
Run model

Workbench

These examples were written for this endpoint. Replace the prompt and sample media URLs with your own accessible inputs, then check the current quote and model availability before submitting.

  • Input types
    Text
    Output types
    Text
    Input
    $9 / 1 million tokens
    Cached input
    $0.90 / 1 million tokens
    Output
    $45 / 1 million tokens
    Context
    1M
    Max output
    128K

    $9/45per 1 million tokens · input / output

    long-form-continuation

    A continuation request with the entire prior text sent as written, no summarising layer. This is Fable 5's home ground: consistency comes from the context, not from reminders. max_tokens is raised to 16,000 because the ask is a whole passage rather than an opening.

    • 16000
  • Input types
    Text
    Output types
    Text
    Input
    $9 / 1 million tokens
    Cached input
    $0.90 / 1 million tokens
    Output
    $45 / 1 million tokens
    Context
    1M
    Max output
    128K

    $9/45per 1 million tokens · input / output

    established-facts

    The character's committed facts listed on their own in the system message, then the model continues. This example exists to show what belongs in the system message and what stays in the verbatim history.

    • 8000
  • Input types
    Text
    Output types
    Text
    Input
    $9 / 1 million tokens
    Cached input
    $0.90 / 1 million tokens
    Output
    $45 / 1 million tokens
    Context
    1M
    Max output
    128K

    $9/45per 1 million tokens · input / output

    scene-from-reference

    A description request in words only: the location notes travel in the message ahead of the ask, since this endpoint takes text, and the model renders them in the established voice.

    • 4096

This example starts at Input $9 / Output $45 per 1 million tokens. The Playground updates the estimate when you change the settings.