LLMCapability
Represents a specific capability or feature of an LLM (Large Language Model). This is a sealed class, where each capability is represented as a subclass or data object.
Inheritors
Types
Represents a specialized capability for audio-related functionalities in the context of a LLM. This capability is used in models that can involving audio processing, such as transcription, audio generation, or audio-based interactions.
Represents the "completion" capability for Language Learning Models (LLMs). This capability typically encompasses the generation of text or content based on the given input context. It belongs to the LLMCapability sealed class hierarchy and is identifiable by the embed ID.
Represents a specific language model capability associated with handling documents.
Represents the capability of generating embeddings within the context of language models.
Represents a capability in the Large Language Model (LLM) for caching.
Represents a structured schema capability for a language model. The schema defines certain characteristics or functionalities related to data interaction and encoding using specific formats.
Represents the capability of the language model to perform speculative responses. This capability allows the model to generate responses with varying degrees of likelihood, often used for exploratory or hypothetical scenarios.
Represents the temperature capability of a language model.
Represents how tools calling can be configured for the LLM.
Represents the capability of tools within the LLM capability hierarchy.
Represents a large language model (LLM) capability associated with vision-based tasks. This capability is typically used in models that can process, analyze, and infer insights from visual data or visual representations.