ai-mocks-openai/dev.mokksy.aimocks.openai.model.chat/ChatCompletionRequest

ChatCompletionRequest

data class ChatCompletionRequest(val messages: List<Message>, val model: String, val store: Boolean = false, val reasoningEffort: String = "medium", val metadata: Metadata? = null, val maxCompletionTokens: Int? = null, val frequencyPenalty: Double? = 0.0, val presencePenalty: Double? = 0.0, val responseFormat: ResponseFormat? = null, val temperature: Double = 1.0, val topP: Double? = 1.0, val n: Int? = 1, val stop: List<String>? = null, val logitBias: Map<String, Int>? = null, val user: String? = null, val seed: Int? = null, val stream: Boolean = false, val streamOptions: ChatCompletionStreamOptions? = null, val tools: List<Tool>? = null, val toolChoice: ToolChoice? = null)(source)

Represents a request for generating a chat-based completions in an OpenAI-like environment. See Create chat completions.

This data class is used for serialization and defines the parameters required to send a chat completions request, including the input messages, model to use, and various tuning parameters.

Author

Konstantin Pavlov

Constructors

ChatCompletionRequest

constructor(messages: List<Message>, model: String, store: Boolean = false, reasoningEffort: String = "medium", metadata: Metadata? = null, maxCompletionTokens: Int? = null, frequencyPenalty: Double? = 0.0, presencePenalty: Double? = 0.0, responseFormat: ResponseFormat? = null, temperature: Double = 1.0, topP: Double? = 1.0, n: Int? = 1, stop: List<String>? = null, logitBias: Map<String, Int>? = null, user: String? = null, seed: Int? = null, stream: Boolean = false, streamOptions: ChatCompletionStreamOptions? = null, tools: List<Tool>? = null, toolChoice: ToolChoice? = null)

Properties

frequencyPenalty

@SerialName(value = "frequency_penalty")

val frequencyPenalty: Double?

The penalty value for repetitive token usage in the response.

@SerialName(value = "logit_bias")

val logitBias: Map<String, Int>?

Modify the likelihood of specified tokens appearing in the completion.

maxCompletionTokens

@SerialName(value = "max_completion_tokens")

val maxCompletionTokens: Int?

The maximum number of tokens allowed in the generated completions.

val messages: List<Message>

A list of input messages that define the conversation context, each with a role and content.

val metadata: Metadata?

Optional metadata associated with the request, such as tags.

val model: String

The identifier of the language model to be used for generating the completions.

val n: Int?

How many chat completion choices to generate for each input message.

presencePenalty

@SerialName(value = "presence_penalty")

val presencePenalty: Double?

The penalty value for new token usage in the response.

reasoningEffort

@SerialName(value = "reasoning_effort")

val reasoningEffort: String

Specifies the level of computational effort to apply during reasoning ("low", "medium", "high").

@SerialName(value = "response_format")

val responseFormat: ResponseFormat?

Defines the response format, including optional JSON schema support.

val seed: Int?

Can be used to produce deterministic responses in testing.

val stop: List<String>?

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

val store: Boolean

A flag indicating whether the conversation context should be stored for further use.

val stream: Boolean

Whether to stream the response.

@SerialName(value = "stream_options")

val streamOptions: ChatCompletionStreamOptions?

Options for streaming responses.

val temperature: Double

A value between 0.0 and 1.0 that controls the randomness of the generated response.

@SerialName(value = "tool_choice")

val toolChoice: ToolChoice?

Controls which (if any) function is called by the model.

val tools: List<Tool>?

List of tools the model may call.

@SerialName(value = "top_p")

val topP: Double?

An alternative to sampling with temperature, called nucleus sampling.

val user: String?

A unique identifier representing your end-user.