Anthropic: Claude Fable Latest

Claude Fable Latest is a text model from Anthropic, published under the model id ~anthropic/claude-fable-latest on api14. It is not a fixed snapshot. As the provider description states, this model always redirects to the latest model in the Claude Fable family, so requests sent to this id are served by whichever Fable release is current at the time of the call. That makes it a convenient choice when you want to stay on the newest Fable model without changing your integration each time Anthropic ships an update.

The model accepts text, images and files as input, and it produces text as output. Its context window is 1,000,000 tokens, which allows very large inputs to be handled in a single request, whether that is a long document, a large codebase excerpt or a lengthy conversation history. The maximum output is 128,000 tokens, so responses can be substantial as well. Supported request parameters include include_reasoning, max_completion_tokens, max_tokens, reasoning, reasoning_effort, response_format, stop, structured_outputs, tool_choice, tools and verbosity.

Those capabilities point to a few practical uses. The large context window suits document analysis, summarisation and question answering over long files, especially since files and images can be passed in directly alongside text. Tool calling and tool_choice make it a reasonable fit for agent style workflows where the model needs to invoke functions or external services. Structured outputs and response_format help when you need JSON or another machine readable shape rather than free prose. The reasoning and reasoning_effort parameters let you trade off how much internal reasoning the model performs, which is useful for tasks that range from quick extraction to multi step problem solving.

Because api14 exposes an OpenAI-compatible API, you call this model the same way you would call an OpenAI model, using the model id ~anthropic/claude-fable-latest in your request. You can send the parameters listed above in the request body, including tools and tool_choice for function calling, response_format or structured_outputs for constrained output, and stop sequences to control where generation ends. The include_reasoning, reasoning and reasoning_effort parameters give you control over reasoning behaviour, while max_tokens and max_completion_tokens cap the response length and verbosity adjusts how verbose the output is.

One thing to keep in mind is that this id is a moving target. Since it always redirects to the latest Fable model, behaviour can change when Anthropic updates the family, which may matter if you need reproducible results across time. In that case a pinned model version would fit better. Also, output is text only, so if you need image or audio generation, a different model is required.