Struct ModelConfig 

pub struct ModelConfig {

    pub id: String,
    pub name: String,
    pub api: ApiProtocol,
    pub provider: String,
    pub base_url: String,
    pub api_key: String,
    pub reasoning: bool,
    pub context_window: u32,
    pub max_tokens: u32,
    pub cost: CostConfig,
    pub headers: HashMap<String, String>,
    pub compat: Option<OpenAiCompat>,
    pub credentials: Option<Arc<dyn CredentialProvider>>,
}

Full model configuration. Knows everything needed to make API calls.

Fields

id: String

Model identifier sent to the API (e.g. “gpt-4o”, “claude-sonnet-4-20250514”).

name: String

Human-friendly name.

api: ApiProtocol

Which API protocol to use.

provider: String

Provider name (e.g. “openai”, “anthropic”, “xai”).

base_url: String

Base URL for API requests (without trailing slash).

api_key: String

Authentication credential for this provider (API key, Bearer token, or access_key:secret[:session_token] for Bedrock). Defaults to an empty string so config files can omit it and supply via env instead.

reasoning: bool

Whether this model supports reasoning/thinking.

context_window: u32

Context window size in tokens.

max_tokens: u32

Default max output tokens.

cost: CostConfig

Cost configuration.

headers: HashMap<String, String>

Additional headers to send with requests.

compat: Option<OpenAiCompat>

OpenAI-compat quirk flags (only for OpenAiCompletions protocol).

credentials: Option<Arc<dyn CredentialProvider>>

Optional refreshable credential source. When Some, every stream() call resolves the API key via credentials.current().await instead of reading api_key directly; the retry loop calls credentials.invalidate().await once on ProviderError::Auth and retries the call before propagating. When None (the default), api_key is used verbatim, preserving the legacy static-key behaviour.

Implementations

impl ModelConfig

pub fn anthropic( id: impl Into<String>, name: impl Into<String>, api_key: impl Into<String>, ) -> Self

Create a new Anthropic model config.

pub fn openai( id: impl Into<String>, name: impl Into<String>, api_key: impl Into<String>, ) -> Self

Create a new OpenAI model config.

pub fn local( base_url: impl Into<String>, model_id: impl Into<String>, api_key: impl Into<String>, ) -> Self

Create a config for a local OpenAI-compatible server (LM Studio, Ollama, etc.). Pass an empty string for api_key — most local servers don’t require authentication.

pub fn google( id: impl Into<String>, name: impl Into<String>, api_key: impl Into<String>, ) -> Self

Create a new Google Generative AI (Gemini) model config.

pub fn openrouter( model_id: impl Into<String>, api_key: impl Into<String>, ) -> Self

Create a new OpenRouter model config. model_id uses the provider/model format (e.g. "anthropic/claude-sonnet-4").

pub fn with_credentials(self, creds: Arc<dyn CredentialProvider>) -> Self

Attach a refreshable credential source. When set, the API key is resolved per-call via credentials.current().await instead of being read directly from self.api_key. The retry loop also calls credentials.invalidate() once on ProviderError::Auth and re-tries the stream call.

pub async fn resolve_api_key(&self) -> Result<String, ProviderError>

Resolve the API key for an outgoing request.

When credentials is set, delegate to its current() method (which may re-fetch from a metadata service or return a cached value). Otherwise fall back to the static api_key field. Providers should call this once at the top of stream() instead of reading api_key directly.

pub async fn invalidate_credentials(&self) -> Result<(), ProviderError>

Signal that the currently cached credential has been rejected. If credentials is set, delegate to its invalidate(); otherwise a no-op. Invoked by the streaming retry loop on ProviderError::Auth.

Trait Implementations

impl Clone for ModelConfig

fn clone(&self) -> ModelConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ModelConfig

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for ModelConfig

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for ModelConfig

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

where __S: Serializer,

Serialize this value into the given Serde serializer.