Documentation

Driver

Defines the required methods for Ai Model Drivers, which are attached to Pipelines via dependency injection. The default driver is the WebRequest GuzzleDriver

Table of Contents

Methods

getModel()  : string
Retrieves the current model identifier.
getPrompt()  : mixed
Retrieves the current prompt.
getSystemPrompt()  : mixed
Retrieves the current system prompt.
output()  : mixed
Executes the API request and returns the response body.
setModel()  : self
Sets the model identifier for the driver.
setPrompt()  : self
Sets the prompt for the driver.
setSystemPrompt()  : self
Sets the system prompt.
stream()  : Generator
Executes a streaming API request.

Methods

getModel()

Retrieves the current model identifier.

public getModel() : string
Return values
string

The model identifier.

getPrompt()

Retrieves the current prompt.

public getPrompt() : mixed
Return values
mixed

The current prompt value.

getSystemPrompt()

Retrieves the current system prompt.

public getSystemPrompt() : mixed
Return values
mixed

The system prompt.

output()

Executes the API request and returns the response body.

public output() : mixed

This method should internally call the connect() method and handle any necessary request execution logic.

Return values
mixed

The response body.

setModel()

Sets the model identifier for the driver.

public setModel(mixed $model) : self
Parameters
$model : mixed

The model identifier.

Return values
self

setPrompt()

Sets the prompt for the driver.

public setPrompt(mixed $prompt) : self
Parameters
$prompt : mixed

The prompt value.

Return values
self

setSystemPrompt()

Sets the system prompt.

public setSystemPrompt(mixed $system) : self

This can be used to provide context or instructions to the AI model.

Parameters
$system : mixed

The system prompt.

Return values
self

stream()

Executes a streaming API request.

public stream() : Generator

Enables streaming mode and returns a generator that yields JSON-encoded data chunks from the API response. This is useful for handling responses that include large or streaming payloads.

Return values
Generator

Yields JSON-encoded data chunks.


        
On this page

Search results