PromptExecutorAPI

Functions

expect abstract fun close()

open suspend fun execute(prompt: Prompt, model: ResolvedModel, tools: List<ToolDescriptor> = emptyList()): Message.Assistant

Executes a given prompt against an already-resolved model and returns the assistant response.

abstract suspend fun execute(prompt: Prompt, model: LLModel, tools: List<ToolDescriptor> = emptyList()): Message.Assistant

Executes a given prompt using the specified LLM and tools, returning a list of responses from the model.

Receives multiple independent choices from an already-resolved model.

Receives multiple independent choices from the LLM. The method is implemented only for some specific providers which support multiple LLM choices.

Executes a given prompt against an already-resolved model and returns its response as a stream.

Executes a given prompt using the specified LLM and returns a stream of output as a flow of StreamFrame objects.

Basic JSON schema generator required for the given model. Return BasicJsonSchemaGenerator by default.

Standard JSON schema generator required for the given model. Return StandardJsonSchemaGenerator by default.

open suspend fun models(): List<LLModel>

Retrieves a list of available models from all LLM clients managed by this executor.

Moderates the content of a prompt using an already-resolved model.

abstract suspend fun moderate(prompt: Prompt, model: LLModel): ModerationResult

Moderates the content of a given message with attachments using a specified LLM.

Resolves the requested model to a ResolvedModel to actually use for the given promptExecutorOperation.