Struct PromptMessage 

pub struct PromptMessage {
    pub role: String,
    pub content: String,
    pub reasoning_content: Option<String>,
    pub name: Option<String>,
    pub tool_calls: Option<Vec<PromptToolCall>>,
    pub tool_call_id: Option<String>,
    pub function_call: Option<ChatFunctionCall>,
}

Common prompt-message shape used by both CLI run and OpenAI serve.

Fields

role: String

content: String

reasoning_content: Option<String>

name: Option<String>

tool_calls: Option<Vec<PromptToolCall>>

tool_call_id: Option<String>

function_call: Option<ChatFunctionCall>

Implementations

impl PromptMessage

pub fn new(role: impl Into<String>, content: impl Into<String>) -> Self

Content is passed to the chat template verbatim — including any <think>...</think> blocks in assistant history. Whether reasoning is kept or stripped from history is a per-template policy (DeepSeek strips it, Qwen3-Coder keeps it); pre-splitting here diverged from what transformers.apply_chat_template feeds the same template. reasoning_content is only set when the client supplies reasoning as a separate field.

Trait Implementations

impl Clone for PromptMessage

fn clone(&self) -> PromptMessage

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for PromptMessage

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

Formats the value using the given formatter.

impl Serialize for PromptMessage

fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>

where S: Serializer,

Serialize this value into the given Serde serializer.