Struct OpenAIConfig 

pub struct OpenAIConfig {
    pub websocket_mode: bool,
    pub responses_store: Option<bool>,
    pub responses_include: Vec<String>,
    pub service_tier: Option<OpenAIServiceTier>,
    pub manual_compaction: OpenAIManualCompactionConfig,
    pub hosted_shell: OpenAIHostedShellConfig,
    pub tool_search: OpenAIToolSearchConfig,
}

OpenAI-specific provider configuration

Fields

websocket_mode: bool

Enable persistent Responses API WebSocket generation and streaming on native OpenAI and OpenAI-compatible Responses endpoints. This is an opt-in path designed for long-running, tool-heavy workflows. A persistent connection removes per-request HTTP handshake overhead, so pair it with service_tier = "ultrafast" on gpt-6-astra/gpt-6.1-sol for agentic tool-call bursts (Ultrafast docs strongly recommend WebSockets). Streaming falls back to HTTP only before any output event is exposed.

responses_store: Option<bool>

Optional Responses API store flag. Set to false to avoid server-side storage when using Responses-compatible models.

responses_include: Vec<String>

Optional Responses API include selectors. Example: ["reasoning.encrypted_content"] for encrypted reasoning continuity.

service_tier: Option<OpenAIServiceTier>

Optional native OpenAI service_tier request parameter. Leave unset to inherit the Project-level default service tier. Options: “flex”, “priority”, “ultrafast”. “ultrafast” is the fastest tier: GA for gpt-6-astra and gpt-6.1-sol, preview-only for gpt-5.6-sol (contact your OpenAI account team). It costs more, starts at low TPM limits (Astra 500k/1M/5M, 6.1-sol 1M/4M/40M), supports US/global processing on Astra and US/EU/global on 6.1-sol, and pairs best with websocket_mode = true.

manual_compaction: OpenAIManualCompactionConfig

Manual /compact defaults for the native OpenAI standalone compaction endpoint.

hosted_shell: OpenAIHostedShellConfig

Optional hosted shell configuration for OpenAI native Responses models.

tool_search: OpenAIToolSearchConfig

Hosted tool search configuration for OpenAI Responses-compatible models.

Trait Implementations

impl Clone for OpenAIConfig

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for OpenAIConfig

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

Formats the value using the given formatter.

impl Default for OpenAIConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for OpenAIConfig

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl JsonSchema for OpenAIConfig

fn schema_name() -> Cow<'static, str>

The name of the generated JSON Schema.

fn schema_id() -> Cow<'static, str>

Returns a string that uniquely identifies the schema produced by this type.

fn json_schema(generator: &mut SchemaGenerator) -> Schema

Generates a JSON Schema for this type.

fn inline_schema() -> bool

Whether JSON Schemas generated for this type should be included directly in parent schemas, rather than being re-used where possible using the $ref keyword.

impl Serialize for OpenAIConfig

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

where __S: Serializer,

Serialize this value into the given Serde serializer.