Claude Fable 5

Claude Fable 5 is the next generation of Anthropic’s frontier intelligence, built for the hardest knowledge work and coding problems. It uses adaptive thinking by default, deciding when and how much to reason based on the complexity of each request.

What it’s good at

• Hard coding problems. Long-running coding agents, complex refactors, multi-file changes, and tasks that benefit from deeper reasoning before producing code.
• Knowledge work. Dense documents, research synthesis, technical writing, and analysis that requires holding lots of context and connecting it carefully.
• Vision tasks. Reading diagrams, charts, screenshots, and documents alongside the text prompt.
• Long-context reasoning. Holds large amounts of context coherently across long workflows.

Inputs

• prompt - the user message.
• system_prompt - optional system prompt.
• image - optional input image. Images are priced as (width px * height px) / 750 input tokens.
• max_tokens - maximum number of output tokens, up to 128,000. Adaptive thinking shares this budget with the final response.
• max_image_resolution - caps input image resolution in megapixels to save time and money. Defaults to 0.5 MP.

Pricing

Fable 5 is priced at $10 per million input tokens and $50 per million output tokens. See the Replicate pricing on the right of this page for details.

Links

• Anthropic
• Claude API model IDs and versioning
• Adaptive thinking