Indicate whether the streaming API shall be used.
The identifier of the model which will be shown in the UI.
Maximum number of retry attempts when a request fails. Default is 3.
OptionalmaxMaximum number of tokens to generate. Default is 4096.
The model ID as used by the Anthropic API.
OptionalreasoningWhich Anthropic reasoning API shape to use. Required when reasoningSupport is set.
'effort': adaptive thinking (thinking: { type: 'adaptive' } + output_config: { effort })'budget': extended thinking (thinking: { type: 'enabled', budget_tokens: N })OptionalreasoningWhen set, the UI exposes a reasoning selector and requests are translated to reasoningApi.
OptionalsupportsTrue on models that accept the Anthropic xhigh effort value.
OptionalurlThe Anthropic API compatible endpoint where the model is hosted. If not provided the default Anthropic endpoint will be used.
Indicate whether the model supports prompt caching.
The key for the model. If 'true' is provided the global Anthropic API key will be used.