Struct LlmConfig 

pub struct LlmConfig { /* private fields */ }

Reusable LLM configuration. Each agent that needs an LLM holds its own LlmConfig and calls LlmConfig::request to start a chat request.

Build one with LlmConfig::builder for explicit settings, or with LlmConfig::from_env to read from AGENT_LINE_* environment variables. Multiple agents can share one config or each hold their own (cheap fast model for one step, strong reasoning model for another).

Implementations

impl LlmConfig

pub fn builder() -> LlmConfigBuilder

Start building an explicit LLM configuration.

pub fn from_env() -> Self

Build an LLM configuration from AGENT_LINE_* environment variables.

Reads AGENT_LINE_PROVIDER, AGENT_LINE_LLM_URL, AGENT_LINE_MODEL, AGENT_LINE_API_KEY, AGENT_LINE_NUM_CTX (Ollama context window), and AGENT_LINE_MAX_TOKENS (OpenAI/Anthropic response cap; falls back to AGENT_LINE_NUM_CTX if unset). Defaults to a local Ollama configuration when nothing is set.

If AGENT_LINE_DEBUG is set, the resolved config is logged to stderr once.

pub fn with_model(self, model: impl Into<String>) -> Self

Return a copy of this config with a different model name. All other fields (provider, base URL, API key, token budgets) are preserved.

pub fn request(&self) -> LlmRequestBuilder

Start building an LLM chat request that uses this config.

Each call creates a fresh LlmRequestBuilder; chain .system(), .user(), and .send() on the result. The config itself is not consumed, so an agent can call self.llm.request() repeatedly.

Trait Implementations

impl Clone for LlmConfig

fn clone(&self) -> LlmConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for LlmConfig

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

Formats the value using the given formatter.

impl PartialEq for LlmConfig

fn eq(&self, other: &LlmConfig) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Eq for LlmConfig

impl StructuralPartialEq for LlmConfig