Struct ProviderConfig 

pub struct ProviderConfig {

    pub api_key: Option<String>,
    pub azure: Option<ProviderConfigAzure>,
    pub base_url: String,
    pub bearer_token: Option<String>,
    pub has_bearer_token_provider: Option<bool>,
    pub headers: Option<HashMap<String, String>>,
    pub max_context_window_tokens: Option<f64>,
    pub max_output_tokens: Option<f64>,
    pub max_prompt_tokens: Option<f64>,
    pub model_id: Option<String>,
    pub transport: Option<ProviderConfigTransport>,
    pub type: Option<ProviderConfigType>,
    pub wire_api: Option<ProviderConfigWireApi>,
    pub wire_model: Option<String>,
}

Custom model-provider configuration (BYOK).

Experimental. This type is part of an experimental wire-protocol surface and may change or be removed in future SDK or CLI releases.

Fields

api_key: Option<String>

API key. Optional for local providers like Ollama.

azure: Option<ProviderConfigAzure>

Azure-specific provider options.

base_url: String

API endpoint URL.

bearer_token: Option<String>

Bearer token for authentication. Sets the Authorization header directly. Takes precedence over apiKey when both are set.

has_bearer_token_provider: Option<bool>

When true, the SDK client supplies bearer tokens on demand: the runtime calls the client-session providerToken.getToken callback before each request and applies the returned token as an Authorization: Bearer <token> header. This is the bearer/OAuth scheme used by Azure AD / managed-identity tokens and provider OAuth access tokens (including Anthropic’s), not a provider-specific API-key header such as Anthropic’s x-api-key. The token-acquiring function itself stays on the SDK side and is never serialized; only this flag crosses the wire. When set alongside apiKey/bearerToken, the callback takes precedence: the runtime applies the token returned by providerToken.getToken as the Authorization: Bearer header for each request and does not send the static credential.

headers: Option<HashMap<String, String>>

Custom HTTP headers to include in all outbound requests to the provider.

max_context_window_tokens: Option<f64>

Maximum context window tokens for the model.

max_output_tokens: Option<f64>

Maximum output tokens for the model.

max_prompt_tokens: Option<f64>

Maximum prompt/input tokens for the model.

model_id: Option<String>

Well-known model ID used for capability lookup. When set, agent behavior config and token limits are inferred from this model.

transport: Option<ProviderConfigTransport>

Provider transport. Defaults to “http”.

type: Option<ProviderConfigType>

Provider type. Defaults to “openai” for generic OpenAI-compatible APIs.

wire_api: Option<ProviderConfigWireApi>

Wire API format (openai/azure only). Defaults to “completions”.

wire_model: Option<String>

The model identifier sent to the provider API for inference (the “wire” model), as opposed to modelId which is the well-known base.

Trait Implementations

impl Clone for ProviderConfig

fn clone(&self) -> ProviderConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ProviderConfig

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

Formats the value using the given formatter.

impl Default for ProviderConfig

fn default() -> ProviderConfig

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for ProviderConfig

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for ProviderConfig

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

where __S: Serializer,

Serialize this value into the given Serde serializer.