Struct OpenAIChatDriver 

pub struct OpenAIChatDriver { /* private fields */ }

OpenAI Chat Driver using Open Responses API

Production driver for OpenAI using the Open Responses specification (https://www.openresponses.org/). This is the recommended driver for new projects, offering:

• Better performance with reasoning models (o1, o3, GPT-5)
• Provider-agnostic streaming events
• Native agentic loop support
• 40-80% better cache utilization

For backward compatibility with the Chat Completions API, use OpenAICompletionsChatDriver instead.

Example

use everruns_openai::OpenAIChatDriver;

let driver = OpenAIChatDriver::new("your-api-key");

// With custom endpoint
let driver = OpenAIChatDriver::with_base_url(
    "your-api-key",
    "https://api.example.com/v1/responses",
);

Implementations

impl OpenAIChatDriver

pub fn new(api_key: impl Into<String>) -> Self

Create a new driver with the given API key

pub fn with_base_url( api_key: impl Into<String>, api_url: impl Into<String>, ) -> Self

Create a new driver with a custom API URL

pub fn api_url(&self) -> &str

Get the API URL

pub fn uses_custom_url(&self) -> bool

Check if using a custom base URL

pub async fn compact_conversation( &self, request: CompactRequest, ) -> Result<CompactResponse>

Compact a conversation to reduce context size

This method calls the /v1/responses/compact endpoint to compress the conversation history. User messages are kept verbatim, while assistant messages, tool calls, and tool results are replaced by an encrypted compaction item.

Arguments

• request - The compact request containing the model and input items

Returns

Returns a CompactResponse containing the compacted output items. The output can be used directly as input for the next /v1/responses call.

pub fn can_compact(&self) -> bool

Check if this driver supports the compact endpoint

Returns true for OpenAI’s Responses API. Custom endpoints may or may not support compaction.

Trait Implementations

impl ChatDriver for OpenAIChatDriver

fn chat_completion_stream<'life0, 'life1, 'async_trait>( &'life0 self, messages: Vec<LlmMessage>, config: &'life1 LlmCallConfig, ) -> Pin<Box<dyn Future<Output = Result<LlmResponseStream>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Call the LLM with streaming response

fn list_models<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = Result<Option<Vec<DiscoveredModel>>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

List available models from the provider

fn supports_compact(&self) -> bool

Check if this driver supports the compact endpoint

fn supports_parallel_tool_calls(&self, model: &str) -> bool

Whether this driver can express the request-level parallel_tool_calls preference on the wire for model.

fn compact<'life0, 'async_trait>( &'life0 self, request: CompactRequest, ) -> Pin<Box<dyn Future<Output = Result<Option<CompactResponse>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Compact a conversation to reduce context size

fn chat_completion<'life0, 'life1, 'async_trait>( &'life0 self, messages: Vec<LlmMessage>, config: &'life1 LlmCallConfig, ) -> Pin<Box<dyn Future<Output = Result<LlmResponse, AgentLoopError>> + Send + 'async_trait>>

where 'life0: 'async_trait, 'life1: 'async_trait, Self: 'async_trait,

Call the LLM without streaming (convenience method)

fn supports_stateful_responses(&self) -> bool

Whether this driver persists Responses API state and can resolve tool calls that are reachable only through previous_response_id.

fn effective_context_window(&self, _model: &str) -> Option<usize>

Effective context window for model, when the driver has authoritative model metadata that is not represented by Everruns’ built-in profiles.

impl Clone for OpenAIChatDriver

fn clone(&self) -> OpenAIChatDriver

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for OpenAIChatDriver

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

Formats the value using the given formatter.