ai-mocks-anthropic/dev.mokksy.aimocks.anthropic.model/MessageCreateParams

MessageCreateParams

data class MessageCreateParams(val model: String, val messages: List<MessageCreateParams.Message>, val system: List<MessageCreateParams.SystemPrompt>? = null, val maxTokens: Int? = null, val temperature: Double? = null, val topP: Double? = null, val topK: Int? = null, val stream: Boolean? = null, val metadata: MessageCreateParams.Metadata? = null, val tools: List<Map<String, String>>? = null)(source)

Parameters for creating a message with Anthropic's Claude models. Based on the structure of com.anthropic.models.messages.MessageCreateParams

Constructors

MessageCreateParams

constructor(model: String, messages: List<MessageCreateParams.Message>, system: List<MessageCreateParams.SystemPrompt>? = null, maxTokens: Int? = null, temperature: Double? = null, topP: Double? = null, topK: Int? = null, stream: Boolean? = null, metadata: MessageCreateParams.Metadata? = null, tools: List<Map<String, String>>? = null)

Types

sealed interface Content

Content of a message. Can be a simple string or a list of content blocks.

sealed interface ContentBlock

A content block in a multi-modal message.

data class ContentList(val blocks: List<MessageCreateParams.ContentBlock>) : MessageCreateParams.Content

List of content blocks for multi-modal messages.

@SerialName(value = "image")

data class ImageBlock(val type: String = "image", val source: MessageCreateParams.ImageSource) : MessageCreateParams.ContentBlock

Image content block.

data class ImageSource(val type: String, val mediaType: String, val data: String)

Source information for an image.

data class Message(val role: String, val content: MessageCreateParams.Content)

Message object for Claude API.

data class Metadata(val userId: String? = null)

Metadata about the request.

data class SystemPrompt(val text: String, val type: String = "text", val cacheControl: MessageCreateParams.SystemPrompt.CacheControl? = null, val citations: List<MessageCreateParams.SystemPrompt.Citation>? = null)

System prompt for Claude. A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role.

@SerialName(value = "text")

data class TextBlock(val type: String = "text", val text: String) : MessageCreateParams.ContentBlock

Text content block.

@SerialName(value = "text")

data class TextContent(val text: String?) : MessageCreateParams.Content

Text content for a message.

Properties

@SerialName(value = "max_tokens")

val maxTokens: Int?

The maximum number of tokens to generate before stopping.

val messages: List<MessageCreateParams.Message>

Input messages.

val metadata: MessageCreateParams.Metadata?

An object describing metadata about the request.

val model: String

The model that will complete your prompt. See the models page for a list of available models.

val stream: Boolean?

Whether to incrementally stream the response using server-sent events.

@Serializable(with = SystemPromptsSerializer::class)

val system: List<MessageCreateParams.SystemPrompt>?

System prompt. A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role.

val temperature: Double?

Amount of randomness injected into the response. Defaults to 1.0. Ranges from 0.0 to 1.0. Use temperature closer to 0.0 for analytical / multiple choice, and closer to 1.0 for creative and generative tasks.

val tools: List<Map<String, String>>?

@SerialName(value = "top_k")

val topK: Int?

Only sample from the top K options for each subsequent token. Used to remove "long tail" low probability responses. Defaults to -1, which disables it.

@SerialName(value = "top_p")

val topP: Double?

Use nucleus sampling. In nucleus sampling, we compute the cumulative distribution over all the options for each subsequent token in decreasing probability order and cut it off once it reaches a particular probability specified by top_p. Defaults to -1.0, which disables it.