Struct CompletionRequest 

pub struct CompletionRequest {
    pub context: Option<RequestContext>,
    pub model: String,
    pub messages: Vec<ChatMessage>,
    pub tools: Option<Vec<Tool>>,
    pub tool_choice: Option<ToolChoice>,
    pub temperature: f32,
    pub max_tokens: u32,
    pub reasoning_effort: Option<ReasoningEffort>,
    pub provider_attempt_id: Option<String>,
    pub stream: bool,
    pub stream_options: Option<StreamOptions>,
}

A chat-completion request.

Fields

context: Option<RequestContext>

Fresh caller identity for resource authorization; never serialized.

model: String

Model identifier as registered in the model registry.

messages: Vec<ChatMessage>

Conversation messages in request order.

tools: Option<Vec<Tool>>

Tool names referenced by this record.

tool_choice: Option<ToolChoice>

How the model may use the advertised tools.

temperature: f32

Sampling temperature.

max_tokens: u32

Upper bound on prompt plus completion tokens.

reasoning_effort: Option<ReasoningEffort>

Reasoning budget, when the provider supports it.

provider_attempt_id: Option<String>

Durable runtime identity used for provider idempotency and reconciliation.

stream: bool

OpenAI-compatible streaming flag. Defaults to false and is omitted on the wire in that case.

stream_options: Option<StreamOptions>

OpenAI-compatible streaming options. Providers only include the final usage frame when include_usage is requested explicitly.

Implementations

impl CompletionRequest

pub fn new(model: impl Into<String>, messages: Vec<ChatMessage>) -> Self

New request with the same defaults as the Python wrapper (temperature = 0.3, max_tokens = 4096).

pub fn stream(self, enabled: bool) -> Self

Enable or disable streaming.

pub fn temperature(self, t: f32) -> Self

Set the sampling temperature.

pub fn max_tokens(self, n: u32) -> Self

Set the maximum completion tokens.

pub fn reasoning_effort(self, effort: ReasoningEffort) -> Self

Set the reasoning budget.

pub fn tools(self, tools: Vec<Tool>) -> Self

Attach tools. Mirrors the Python default: if a caller adds tools without an explicit choice, we set tool_choice = auto.

pub fn tool_choice(self, choice: ToolChoice) -> Self

Set the tool-choice policy.

Trait Implementations

impl Clone for CompletionRequest

fn clone(&self) -> CompletionRequest

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for CompletionRequest

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for CompletionRequest

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for CompletionRequest

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.