Readonly
capabilitiesVarious features that the model supports such as tool calling or image input.
Optional
Readonly
imageInput?: booleanWhether image input is supported by the model. Common supported images are jpg and png, but each model will vary in supported mimetypes.
Optional
Readonly
toolCalling?: number | booleanWhether tool calling is supported by the model. If a number is provided, that is the maximum number of tools that can be provided in a request to the model.
Optional
Readonly
detailAn optional, human-readable string which will be rendered alongside the model. Useful for distinguishing models of the same name in the UI.
Readonly
familyOpaque family-name of the language model. Values might be gpt-3.5-turbo
, gpt4
, phi2
, or llama
Readonly
idUnique identifier for the language model. Must be unique per provider, but not required to be globally unique.
Readonly
maxThe maximum number of tokens the model can accept as input.
Readonly
maxThe maximum number of tokens the model is capable of producing.
Readonly
nameHuman-readable name of the language model.
Optional
Readonly
tooltipThe tooltip to render when hovering the model. Used to provide more information about the model.
Readonly
versionOpaque version string of the model.
This is used as a lookup value in LanguageModelChatSelector.version
An example is how GPT 4o has multiple versions like 2024-11-20 and 2024-08-06
Represents a language model provided by a
LanguageModelChatProvider
.Stubbed