Skip to main content

BaseChatModel

lc_core::language_models

Trait BaseChatModel 

Source 
pub trait BaseChatModel: BaseLanguageModel<Vec<Message>, LLMResult> {
    // Required methods
    fn chat<'life0, 'async_trait>(
        &'life0 self,
        messages: Vec<Message>,
        config: Option<RunnableConfig>,
    ) -> Pin<Box<dyn Future<Output = Result<LLMResult, Self::Error>> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait;
    fn stream_chat<'life0, 'async_trait>(
        &'life0 self,
        messages: Vec<Message>,
        config: Option<RunnableConfig>,
    ) -> Pin<Box<dyn Future<Output = Result<Pin<Box<dyn Stream<Item = Result<StreamChunk, Self::Error>> + Send>>, Self::Error>> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait;

    // Provided methods
    fn chat_with_system<'life0, 'async_trait>(
        &'life0 self,
        system: String,
        messages: Vec<Message>,
    ) -> Pin<Box<dyn Future<Output = Result<LLMResult, Self::Error>> + Send + 'async_trait>>
       where Self: Sync + 'async_trait,
             'life0: 'async_trait { ... }
    fn bind_tools(
        &self,
        _tools: Vec<ToolDefinition>,
    ) -> Option<Box<dyn BaseChatModel<Error = Self::Error> + Send + Sync>> { ... }
}
Expand description

Base trait for chat models.

Extends BaseLanguageModel for chat scenarios. Accepts message list as input, returns AI message.

Required Methods§

Source

fn chat<'life0, 'async_trait>( &'life0 self, messages: Vec<Message>, config: Option<RunnableConfig>, ) -> Pin<Box<dyn Future<Output = Result<LLMResult, Self::Error>> + Send + 'async_trait>>
where Self: 'async_trait, 'life0: 'async_trait,

Chat with the model.

§Arguments
  • messages - Message list.
  • config - Optional configuration.
§Returns

LLM result.

Source

fn stream_chat<'life0, 'async_trait>( &'life0 self, messages: Vec<Message>, config: Option<RunnableConfig>, ) -> Pin<Box<dyn Future<Output = Result<Pin<Box<dyn Stream<Item = Result<StreamChunk, Self::Error>> + Send>>, Self::Error>> + Send + 'async_trait>>
where Self: 'async_trait, 'life0: 'async_trait,

Stream chat with the model.

§Arguments
  • messages - Message list.
  • config - Optional configuration.
§Returns

Stream of StreamChunk items. Chunks carry the text delta; the final chunk may additionally carry StreamChunk::token_usage when the provider reports usage.

Provided Methods§

Source

fn chat_with_system<'life0, 'async_trait>( &'life0 self, system: String, messages: Vec<Message>, ) -> Pin<Box<dyn Future<Output = Result<LLMResult, Self::Error>> + Send + 'async_trait>>
where Self: Sync + 'async_trait, 'life0: 'async_trait,

Chat with system prompt.

§Arguments
  • system - System prompt.
  • messages - Message list.
§Returns

LLM result.

Source

fn bind_tools( &self, _tools: Vec<ToolDefinition>, ) -> Option<Box<dyn BaseChatModel<Error = Self::Error> + Send + Sync>>

Bind tool definitions for function calling.

Returns Some(model) with the tools attached when the provider supports tool calling; returns None when it does not. The default returns None, signalling a hard capability limit — callers MUST treat None as “this model cannot call tools” and branch accordingly (e.g. fall back to text-only prompting). Providers that support function calling (OpenAI, Ollama) override this.

This is an explicit result, not a silent degrade: None is the honest answer that tool-calling is unavailable on this model.

Dyn Compatibility§

This trait is dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety".

Implementors§