ChatGPT-Java-API/com.cjcrafter.openai.completions/CompletionRequest

CompletionRequest

data class CompletionRequest @JvmOverloads constructor(var model: String, var prompt: Any? = null, var suffix: String? = null, var maxTokens: Int? = null, var temperature: Number? = null, var topP: Number? = null, var n: Int? = null, var stream: Boolean? = null, var logprobs: Int? = null, var echo: Boolean? = null, var stop: Any? = null, var presencePenalty: Number? = null, var frequencyPenalty: Number? = null, var bestOf: Int? = null, var logitBias: Map<String, Int>? = null, var user: String? = null)

Holds the configurable options that can be sent to the OpenAI Completions API. For most use cases, you only need to set model and prompt. For more detailed descriptions for each option, refer to the Completions Wiki.

prompt can be either a singular String, a List<String>, or a String[]. Providing multiple prompts is called batching, and it can be used to reduce rate limit errors. This will cause the CompletionResponse to have multiple choices.

You should not set stream. The stream option is handled using []

Constructors

CompletionRequest

@JvmOverloads

constructor(model: String, prompt: Any? = null, suffix: String? = null, maxTokens: Int? = null, temperature: Number? = null, topP: Number? = null, n: Int? = null, stream: Boolean? = null, logprobs: Int? = null, echo: Boolean? = null, stop: Any? = null, presencePenalty: Number? = null, frequencyPenalty: Number? = null, bestOf: Int? = null, logitBias: Map<String, Int>? = null, user: String? = null)

Create a CompletionRequest instance. Recommend using builder instead.

Types

Builder

class Builder

Builder is a helper class to build a CompletionRequest instance with a stable API. It provides methods for setting the properties fo the CompletionRequest object. The build method returns a new CompletionRequest instance with the specified properties.

Companion

object Companion

Properties

bestOf

var bestOf: Int?

Generates best_of completions server-side and returns the "best" (the one with the highest log probability per token).

echo

var echo: Boolean?

Echo back the prompt in addition to the completion.

frequencyPenalty

var frequencyPenalty: Number?

Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.

logitBias

var logitBias: Map<String, Int>?

Modify the likelihood of specified tokens appearing in the completion.

logprobs

var logprobs: Int?

Include the log probabilities on the logprobs most likely tokens, as well the chosen tokens.

maxTokens

var maxTokens: Int?

The maximum number of tokens to generate in the completion.

model

var model: String

ID of the model to use.

var n: Int?

How many completions to generate for each prompt.

presencePenalty

var presencePenalty: Number?

Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.

prompt

var prompt: Any?

The prompt(s) to generate completions for (either a String, List<String>, or String[])

stop

var stop: Any?

Up to 4 sequences where the API will stop generating further tokens.

stream

var ~~stream~~: Boolean?

Whether to stream back partial progress.

suffix

var suffix: String?

The suffix that comes after a completion of inserted text.

temperature

var temperature: Number?

What sampling temperature to use, between 0 and 2.

topP

var topP: Number?

An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass.

user

var user: String?

A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse.