Struct ModelRequest 

pub struct ModelRequest {
    pub model: String,
    pub instructions: Option<String>,
    pub messages: Vec<ModelMessage>,
    pub thinking: ThinkingConfig,
    pub tools: Vec<ToolDefinition>,
    pub session_id: Option<String>,
}

A request to a model provider containing the conversation, model name, thinking configuration, and available tool definitions.

Fields

model: String

The model identifier to use (e.g. "gpt-5.5", "claude-sonnet-4-20250514").

instructions: Option<String>

Stable base instructions sent in the provider’s instructions field (Responses API) or as the first system message (Chat Completions, Anthropic, Gemini). Kept separate from Self::messages so that dynamic context blocks (developer messages) don’t invalidate the provider’s prompt cache for this prefix.

messages: Vec<ModelMessage>

The conversation messages to send to the model.

thinking: ThinkingConfig

The thinking/reasoning effort level to request.

tools: Vec<ToolDefinition>

Tool definitions the model may invoke.

session_id: Option<String>

Stable session id for provider-side prompt-cache affinity.

Providers such as Charm Hyper use this to set x-session-id and x-session-affinity so consecutive turns of the same agent session hit the same KV-cache shard. Not serialized to disk/transcripts.

Trait Implementations

impl Clone for ModelRequest

fn clone(&self) -> ModelRequest

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ModelRequest

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for ModelRequest

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for ModelRequest

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

where __S: Serializer,

Serialize this value into the given Serde serializer.