Action
    Preparing search index...

    Well-known tags that an administrator can assign to an AI model. Pass them to GetModel(tag, additionalTags) to pick a model by capability. Administrators may also assign arbitrary custom tags (for example a domain such as "oil and gas"); these constants only name the built-in ones. Tags are matched case-insensitively.

    2.6

    Hierarchy (View Summary)

    Index

    Methods

    • Casts this object to the specified type. Throws error on failure.

      Type Parameters

      • T1

        The target type of the cast.

      • T2

        The type of the object to cast.

      Parameters

      Returns T1

    • Determines whether the specified object is equal to the current object.

      Parameters

      • obj: any

        The object to compare with the current object.

      Returns boolean

      true if the specified object is equal to the current object; otherwise, false.

    • Serves as the default hash function.

      Returns number

      A hash code for the current object.

    • Returns a string that represents the current object.

      Returns string

      A string that represents the current object.

    • Casts this object to the specified type. Returns null on failure.

      Type Parameters

      • T1

        The target type of the cast.

      • T2

        The type of the object to cast.

      Parameters

      Returns T1 | null

    • Determines whether the specified object instances are considered equal.

      Parameters

      • objA: any

        The first object to compare.

      • objB: any

        The second object to compare.

      Returns boolean

      true if the objects are considered equal; otherwise, false. If both objA and objB are null, the method returns true.

    Default capability

    Coding: string

    The model has strong code generation and editing capabilities.

    2.6

    Fast: string

    The model is expected to give fast responses, suitable for interactive or hot-path use.

    2.6

    LongContext: string

    The model supports a large context window (128k tokens or more).

    2.6

    Multilingual: string

    The model has strong non-English language support.

    2.6

    Streaming: string

    The model can stream responses; a requirement for driving chat.

    2.6

    Thinking: string

    The model is capable of extended reasoning, suitable for driving agentic flows.

    2.6

    Vision: string

    The model can interpret images; a requirement for most agentic flows.

    2.6