Struct ChatCompletionRequest 

pub struct ChatCompletionRequest {
    pub model: ModelField,
    pub messages: Vec<ChatCompletionMessage>,
    pub stream: bool,
    pub temperature: Option<f32>,
    pub max_tokens: Option<u32>,
    pub strict_capabilities: Option<bool>,
    pub tools: Option<Vec<ToolDefinition>>,
    pub tool_choice: Option<ToolChoice>,
    pub response_format: Option<ResponseFormatRequest>,
    pub top_p: Option<f32>,
    pub stop: Option<StopField>,
}

OpenAI-compatible chat completion request

Accepts either a single model string or an array of model strings for multiplex mode. Each model string may contain a provider prefix (e.g., “copilot:gpt-4o”) parsed by the provider resolver.

Fields

model: ModelField

Model identifier(s) — single string or array for multiplex

messages: Vec<ChatCompletionMessage>

Conversation messages

stream: bool

Whether to stream the response

temperature: Option<f32>

Temperature for response randomness (0.0 - 2.0)

max_tokens: Option<u32>

Maximum tokens to generate

strict_capabilities: Option<bool>

Enable strict capability checking (reject unsupported parameters)

tools: Option<Vec<ToolDefinition>>

Tool definitions for function calling

tool_choice: Option<ToolChoice>

Controls which tools the model may call

response_format: Option<ResponseFormatRequest>

Controls the response format (text, json_object, or json_schema)

top_p: Option<f32>

Nucleus sampling parameter (0.0 - 1.0)

stop: Option<StopField>

Stop sequence(s) that halt generation — single string or array

Trait Implementations

impl Debug for ChatCompletionRequest

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for ChatCompletionRequest

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.