Struct StreamConfig 

pub struct StreamConfig {
    pub model_config: ModelConfig,
    pub system_prompt: String,
    pub messages: Vec<Message>,
    pub tools: Vec<ToolDefinition>,
    pub thinking_level: ThinkingLevel,
    pub max_tokens: Option<u32>,
    pub temperature: Option<f32>,
    pub cache_config: CacheConfig,
    pub response_format: ResponseFormat,
}

Configuration for a streaming LLM call

Fields

model_config: ModelConfig

Complete provider identity: model id, api_key, base_url, compat flags, cost rates. All providers read model_config.id and model_config.api_key; most also read model_config.base_url and model_config.headers.

system_prompt: String

messages: Vec<Message>

tools: Vec<ToolDefinition>

thinking_level: ThinkingLevel

max_tokens: Option<u32>

temperature: Option<f32>

cache_config: CacheConfig

Prompt caching configuration. Default: enabled with auto strategy.

response_format: ResponseFormat

Desired output shape. Text (the default) preserves the historical behaviour; JsonObject / JsonSchema request constrained JSON output from providers that support it natively (OpenAI, Google) or via tool-call emulation (Anthropic). Bedrock surfaces ProviderError::SchemaMismatch when set on a non-Anthropic foundation model that lacks structured-output support. See the capability matrix in docs/specs/developer/provider.md for per-provider coverage.

Trait Implementations

impl Clone for StreamConfig

fn clone(&self) -> StreamConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for StreamConfig

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

Formats the value using the given formatter.